Trial Proceedings Transcript Vol. 5
Public Court Documents
August 12, 1983
187 pages
Cite this item
-
Case Files, McCleskey Legal Records. Trial Proceedings Transcript Vol. 5, 1983. 7d6cfec3-5aa7-ef11-8a69-6045bdd6d628. LDF Archives, Thurgood Marshall Institute. https://ldfrecollection.org/archives/archives-search/archives-item/234e8fc7-80af-4aad-9c55-3ab6ef422b88/trial-proceedings-transcript-vol-5. Accessed April 28, 2025.
Copied!
{
| 247
;
2 IN THE UNITED 3TATES DISTRICT COURT
3 FOR THE NORTHERN DISTRICT OF GEORGIA
a ATLANTA DIVISION
: ig
& |WARREN MCCLESKEY ) DOCKET NO. CB1 2434A
7 ) |
PLAINTIFF, ) ATLANTA, GEORSIA |
: | 5 ) AUGUST 12, 1983
[WALTER D. ZANT, WARDEN, )
13 )
| )
ty DEFENDANT. )
2 BAT
wid VOLUME V
: 14 | TRANSCRIPT OF PROCEEDINGS
: 4
!
ib | JUDGE.
|
i
|
|
ig APPEARANCES OF COUNSEL: i
” |
w 19 FOR THE PLAINTIFF: JOHN CHARLES BOGERY TIMOTHY K. FORD
| AND ROBERT H. STROUP,
20h | |
|
21 |FOR THE DEFENDANT: MARY BETH WESTMORELAND AND SUSAN V.
| BOLEYN. |
22 ae |
23 |
| JIM PUGH ;
ol GRE | OFFICIAL COURT REPORTER
ROOM 23467, 7% SPRING STREET: T.W.
23 ATLANTA, SEORGIA 20303
|
m
i
n
s
aad WITNESSES
DIRECT ©CROSS REDIRECT RECROSS
WITNESSES FOR THE PLAINTIFF:
BALDUS, DAVID C. 834
549
1 DOCUMENTARY EVIDENCE
2 MARKED RECEIVED
3 | DB-86 ome 872
| pB-87 a72 27s
4 DE-8% ars £85
NB=90 881 88%
5 DBE-91 £os S64
® DB-92 ag7 893
& DB-93 394 902
DB-94 896 202
7 DB-95 204 214
DB-96 216 022
8 DB-97 232 938
DB-96A 934 938
9 DB-98 929 9240
DB-99 : 943
10 DB-100 047
DB-101 953
11 DB-102 954
DE-103 261 963
12 DB-104 264 972
DB-10% 973 974
13 DB-1064 978 979
DB-107 280 o81
14 DB-108 . 983 285
DB-109 $92 993
15 DB-110 997 98
DB-111 1000 100%
16 |
17
18
w 19
20
21 |
22
23
24
25 |
|
|
|
830
1 (ATLANTA, FULTON COUNTY, GEORGIA AUGUST 12, 1983,
2 IN OPEN COURT.)
THE COURT: LET ME STATE TO YOU THE OPINION OR THE
THOUGHT THAT I HAVE, AND THEN GET YOUR COMMENTS ON IT.
3
4
S
& BECAUSE OF THE CARE YOU AND MR. FORD HAVE TAKEN IN
7 AUTHENTICATING OR ACCREDITING DOR WHATEVER THE PROFER WORD IS THE
38 EVIDENCE YOU“RE PRESENT ING. WE HAVE SPENT A LOT OF TIME ALREADY
@ AND 1 HAVE NOT BEEN ABLE YET TO FULLY UNDERSTAND THE POSITION OF
10 |THE STATE WITH REFERENCE TO THIS DATA.
i1 I KNOW SOME THINGS THAT THEY FEEL. BUT I DON’T KNOW ALL
12 THE THINGS THAT THEY FEEL, AND PARTICULARLY BEYOND A GENERAL
13. RELEVANCY OBJECTION WHICH IS, ALL THIS IS REAL INTERESTING, BUT
14 IT DOESN‘T SHOW THAT MCCLESKEY WAS DISCRIMINATED AGAINST.
15 AND I AM UNABLE TO REALLY APPRECIATE AND YOU MAY BE,
16 TOO, THE MAJOR PROBLEMS THAT THEY FEEL ARE IN THE STATISTICAL
17 STUDY.
18 I THINK IT WOULD PROBABLY BE MOST EFFICIENT AND ORDERLY
Ww 19% IF YOU WOULD FINISH WITH THIS WITNESS, FINISH WITH THE
20 STATISTICIAN THAT COLLABORATED WITH HIM, AND NOT PUT ON YOUR
21 SECOND EXPERT —- I DON’T KNOW IF YOU HAVE ANYTHING ELSE. BESIDE
22 PROFESSOR BALDUS AND THE STATISTICIAN -—- AND THEN LET“5 HEAR
23 WHAT THE STATE HAS TO SAY ABOUT THE VALIDITY OF THE STUDY; SAVE
24 YOUR SECOND EXPERT AS A REBUTTAL WITNESS S0 THAT WE CAN DEAL
23 WITH HIM ONLY ON THOSE THINGS WHICH ARE IN ISSUE.
po
ts
MR. BOGER: LET ME CONFER WITH CO-COUNSEL FOR A MOMENT,
[ YOUR HONOR.
YOUR HONOR. THAT SEEMS ACCEPTABLE TO US.
THE ONE QUESTION I WOULD HAVE. AS YOU KNOW, PROFESSOR
BERK HAS BEEN UNDER THE RULE. IF PROFESSOR BERK IS TO BE A
REBUTTAL WITNESS, ALTHOUGH WE DON’T KNOW WHEN HE WOULD BE HERE,
IF THE STATE-"S CASE GOES FORWARD, WE WONDER WHETHER THERE WOULD
BE ANY PROBLEM UNDER THOSE CIRCUMSTANCES. HAVING HIM EXEMPTED
FROM THE RULE SO HE CAN HEAR WHAT THE STATE HAS TQ SAY AND
o
R
TH
R
E
E
R
R
en
B
E
S
T
TESTIFY AS A REBUTTAL WITNESS.
11 THE COURT: THE FROPOSITION HAS SOME APPEAL. MS.
12 WESTMORELAND. WHAT SAY YE.
13 MS. WESTMORELAND: YOUR HONOR, I WOULD STILL HAVE THE
14 SAME CONCERNS THAT 1 HAVE ABOUT THE WITNESS REMAINING IN THE
15 COURTROOM TO HEAR OUR ENTIRE TESTIMONY, BECAUSE I UNDERSTAND HIS
146 TESTIMONY IS BASICALLY IN RELATION TO HIS REVIEW OF PROFESSOR
17 BALDUS” REPORT AND HIS CONCLUSIONS.
13 THE COURT: WHAT IS HIS EXPERTISE?
oo 19 MR. BOGER: YOUR HONOR, MR. FORD COULD SPEAK TO THAT.
20 MR. FORD: YOUR HONOR, PROFESSOR BERK IS A FROFESS0OR. I
21 BELIEVE, IN THE DEPARTMENT OF SOCIOLOGY, UNIVERSITY OF
22 CALIFORNIA AT SANTA BARBARA.
23 | IN BRIEF HE IS. AMONG OTHER THINGS. ONE OF THE TWELVE
24 ~~ |PEOPLE WHO WAS ON A NATIONAL ACADEMY CF SCIENCE REVIEW BOARD OF
23 ALL THE LITERATURE RELATIVE TO RACIAL DISCRIMINATION IN
852
1 SENTENCING. | ME‘S ALSO VERY EXPERIENCED IN THIS KIND OF DATA
| GATHERING AND ANALYSIS.
HE IS FAMILIAR AT THIS POINT WITH PROFESSOR BALDUS
& WORK, AND UF TO AND INCLUDING HIS MOST RECENT REPORTS WHICH ARE
BEFORE THE COURT AND COMING IN BEFORE THE COURT, AS WELL AS
~~
oO
DOCTOR KATZ REPORT CRITICIZING THOSE, AND AS THE COURT HAS
INDICATED OR AS MR. BOGER”S INDICATED. WE HAVE KEPT HIM UNDER
L
I
THE RULE PURSUANT TO THE COURT’S SUGGESTION REGARDING ANY
10 INFORMATION AS TO WHAT HAS TRANSPIRED IN COURT.
11 BUT IT WAS OTHER ANTICIPATION THAT BECAUSE I THINK
12 DIFFERENT CONSIDERATIONS THE COURT MAY APPLY TO DOCTOR KATZI® -——
13 THE COURT: YOU HAVEN‘T ANSWERED MY QUESTION, COUNSEL.
14 MR. FORD: I-“M SORRY,
13 | THE COURT: WHAT IS HIS EXPERTISE AND HOW IS IT GOING TO
16 BE USED IN THIS CASE?
17 MR. FORD: HIS EXPERTISE IS BOTH WITH REGARD TO THE
ig SPECIFIC STATISTICAL AND DATA GATHERING TECHNIQUES THAT HAVE
@ 12 BEEN UTILIZED HERE: THE ANALYTICAL TECHNIQUES AND THE BROAD
20 PERSPECTIVE OF THE QUALITY OF THIS STUDY WITH REGARD TO THE
21 TOTAL PICTURE, AND WITH REGARD TO, INSOFAR AS IT'S A REBUTTAL
22 WITNESS, 1 THINK WE WILL SHOW THAT HE IS PARTICULARLY QUALIFIED
23 TO ADDRESS SOME OF THE SPECIFIC ANALYTICAL CONCERNS THAT HAVE
24 BEEN RAISED BY DOCTOR KATZ” REPORT.
25 THE COURT: HE HAS NOT INDEPENDENTLY PERFORMED ANY OF
4
ff
B
N
MANIPULATION OF THE DATA?
MR. FORD: THAT-S CORRECT. YOUR HONOR. MY
UNDERSTANDING, HE HAS REVIEWED —-
THE COURT! AND IS HERE PRIMARILY TO VALIDATE THE
METHODOLOGY?
MR. FORD: THATS CORRECT, YOUR HONOR.
MR. BOGER: YOUR HONOR, -- | |
MR. FORD: AND TO GIVE THE COURT SOME PERSPECTIVE AS TO
WHERE THIS FITS IN THE TOTAL SCHEME OF POSSIBLE THINGS THAT CAN
BE DONE AND HAS BEEN DONE IN THIS KIND OF RESEARCH.
MR. BOGER: YOUR HONOR, 1 BELIEVE 1 NEED TO TALK A
LITTLE BIT FURTHER WITH MR. FORD ABOUT THIS. BUT THAT PROFESSOR
BERK MAY AT SOME POINT LOOKED AT SOME OF THE DATA. AND I DON‘T
KNOW THE EXTENT OF THAT, BUT WE MIGHT WANT TO CONFER A LITTLE
MORE ON THAT BEFORE GIVING YOU A FINAL ANSWER.
THE COURT: AFTER FOUR DAYS OF TESTIMONY, WHERE WE ARE
IS THAT A STATE-OF-THE-ART STUDY HAS BEEN DONE WHICH SHOWS THAT
RACE OF THE VICTIM HAS SOME EFFECT. ALTHOUGH IT DOES NOT OPERATE
UNIFORMLY THROUGHOUT THE SYSTEM AND IS NOT THE SOLE EFFECT IN
PREDICTING WHO MIGHT GET THE DEATH PENALTY. THAT’S WHAT WE-VE
GOT AFTER FOUR DAYS, AND I THINK AT THIS POINT IN TIME. IT MOVES
FROM THE SOCIAL SCIENCE FIELD INTO THE LEGAL FIELD.
1/M JUST GIVING YOU AN IMPRESSION. HAVING ONLY HEARD
YOUR SIDE OF THE STORY. AND HAVING NOT EVEN HEARD ALL OF
THAT.
RALDUS - DIRECT
1 I THINK I WOULD BE INCLINED TO LET HIM OUT FROM UNDER
2 THE RULE, AND RESTRICT YOU TO THE USE OF HIM SOLELY AS A
3 REBUTTAL EXPERT.
4 MR. BOGER: ALL RIGHT, YOUR HONOR.
. MR. FORD: THANK YOU.
A 6 THE COURT: WHICH MEANS THAT HE REBUTS WHAT THE STATE
3 SAYS.
8 MR. BOGER? FINE, YOUR HONOR. I THINK THAT’S
9 ACCEPTABLE.
10 THE COURT: ALL RIGHT. GO AHEAD.
11 MR. FORD: IF THE COURT PLEASE, WITH REGARD TO MY
12 PREVIOUS REMARKS, MAY I BE HEARD FOR JUST A MOMENT? I HAVE NOT
12 AT THIS TIME GOTTEN TO THE POINT OF FULLY PREPARING PROFESSOR
14 |BERK, AND IF I MISSTATE ANYTHING ABOUT WHAT HE ULTIMATELY
15 |TESTIFIES TO, IT“S SIMPLY MY LACK OF PREPARATION AT THIS POINT.
16 | eel
17 DAVID C. BALDUS,
13 BEING PREVIOUSLY DULY SWORN, RESUMED THE WITNESS STAND AND
ww 19 |TESTIFIED FURTHER AS FOLLOWS:
20 | DIRECT EXAMINATION (CONTD)
21 |BY MR. BOGER:
22 A. PROFESSOR BALDUS, I BELIEVE AT THE CLOSE OF THE TESTIMONY
23 YESTERDAY. YOU HAD BEGUN TO EXPLAIN WHY YOU HAD FOCUSED IN ON
24 CERTAIN SUB-CATEGORIES OF CASES WITHIN YOUR DATA BASE CASES, I
25 BELIEVE YOU TESTIFIED, IN WHICH THE DEATH SENTENCING RATES WERE
BALDUS ~ DIRECT
1 HIGHER THAN IN SOME OTHER CATEGORIES. AND MORE SFECIFICALLY
2 THAN THAT, ON THE B2 AND B7 AGGRAVATING CIRCUMSTANCE CASES.
WOULD YOU PLEASE BRIEFLY REVIEW FOR US THE SIGNIFICANCE
OF DB-85 IN THAT REGARD?
3
4
5 A. YES. THE PURPOSE OF THE PARTICULAR FOCUS ON B7 AND B2 CASES
& WAS THAT THOSE CASES ACCOUNT FOR A VERY HIGH PROPORTION OF ALL
7 THE DEATH SENTENCES THAT ARE IMPOSED IN THE GEORGIA DEATH
8 SENTENCING SYSTEM.
2 I WAS PARTICULARLY CONCERNED ABOUT HOW CASES WERE
10 TREATED WITHIN THAT CATEGORY. AND TABLE 34, WHICH IS FROM THE
11 REPORT, WHICH IS EXHIBIT DB-85. PRESENTS AN OVERVIEW OF THE
32 RESULTS OF THE STATISTICAL ANALYSES THAT WERE DONE ON THE CASES
13 FALLING IN THE B2 AND B7 CATEGORIES.
14 AND WHAT IS APPARENT IS THAT WITHIN THE B2 CASES, WE
13 SEE RACE OF VICTIM EFFECTS. WE SEE THEM ALSO IN THE B7
16 CATEGORY; AND WE SEE, AS WE WILL, STRONG RACE OF DEFENDANT
17 EFFECTS IN THE B7 CATEGORY OF CASES.
18 @. WE THEN TURNED TO DB-8&, MARKED FOR IDENTIFICATION, AND YOU
19 WERE GOING TO EXPLAIN TO US HOW YOU HAD DONE SOME
20 SUBCATEGORIZATION, I BELIEVE, WITHIN THE B2 GROUP OF CASES.
21 CAN YOU CONTINUE THAT ANALYSIS?
= A. YES. THE, THE METHOD WE USED TO COLLECT DATA INVOLVED THE
23 PREPARATION OF A DETAILED SUMMARY OF EACH CASE, AND IT FROVIDED
24 AN OPPORTUNITY TO ANALYZE IN A MORE FINE-GRAIN FASHION. USING
23 QUALITATIVE JUDGMENTS AS WELL AS SIMPLY STATISTICAL JUDGMENTS
356
BALDUS - DIRECT
ABOUT THE FACTORS THAT SHOULD BE USED TO CONTROL FOR THE FACTORS
THAT MIGHT AFFECT THE DEATH SENTENCING DECISIONS.
'@. HOW DID YOU BREAK THOSE CASES QUT?
A. WE STARTED IN THIS ANALYSIS. I WOULD SAY WE CONDUCTED THIS
ANALYSIS WITH RESPECT TO FOUR OR FIVE OF THE SPECIFIC STATUTORY
AGGRAVATING FACTORS.
| THEY WERE PARTICULARLY IMPORTANT WITH REGARD TO
STATUTORY AGGRAVATING FACTORS WHERE THE SAMPLE SIZE WAS SO SMALL
THAT A STATISTICAL ANALYSIS WAS NOT POSSIBLE, FOR THE PURPOSES
OF CONTROLLING FOR BACKGROUND FACTORS.
WE FOCUSED HERE ON A LARGER POOL OF CASES, BECAUSE
THESE ARE EXTREMELY IMPORTANT IN TERMS OF DETERMINING WHO
RECEIVES A DEATH SENTENCE.
@. BY LARGER POOL YOU MEAN?
A. THE B2 CASES.
@. ALL RIGHT.
A. WE PULLED THE B2 CASES OUT OF THE CHARGING AND SENTENCING
STUDY THAT WE HAD IDENTIFIED AND WE PULLED ALL THE B2 CASES OUT
OF THE PROCEDURAL REFORM STUDY. AND AS A CONSEQUENCE, WE WERE
ABLE TO OBTAIN A SAMPLE OF 438 CASES.
WE CALCULATED RACIAL EFFECTS WITHIN THAT GROUP, AND WE
SAW THEM. THEY ARE PRESENTED IN ROW 1 OF DB-8&, SHOWS A, A
STRONG RACIAL DISPARITY WHICH 1S. HAS A HIGH LEVEL OF
STATISTICAL SIGNIFICANCE.
HOWEVER, THERE IS A LOT OF VARIETY AMONG THE B2 CASES,
pe
10
i1
G
U
E
T
E
N
TR
E
E
R
T
R
a
O
E
0 a ~d
| BALDUS - DIRECT
50 WE SOUGHT TO CONTROL FOR THE FURTHER DIMENSIONS OF THOSE
PARTICULAR CASES.
AND WE DID SO FIRST BY SORTING OUT THE ARMED ROBBERY
CASES, AND THOSE ARE REPRESENTED IN ROW 2. PART 2, RATHER. OF
DB-86. AND YOU CAN SEE THAT WHEN YOU LOOK AT ALL ARMED ROBBERY
CASES ALONE, THAT THE STRONG RACIAL EFFECT WITH RESPECT TO THE
RACE OF THE VICTIM PERSISTS. THERES A 29 PERCENTAGE POINT
DIFFERENCE IN THE RATE AT WHICH DEFENDANTS INVOLVED IN ARMED
ROBBERY RECEIVE DEATH SENTENCES BETWEEN WHITE AND BLACK VICTIM
CASES.
THE COURT: WHAT ARE YDU CONTROLLING FOR HERE?
THE WITNESS: JUST ARMED ROBBERY. YOUR HONOR.
THEN WE CONTINUED TO SUB~CATEGORIZE ARMED ROBBERY.
| DOWN UNDER PART 2, Bl, WE SORT THE ARMED ROBBERY CASES INTO
FURTHER SUB-CATEGORIES. THE FIRST CATEGORY IS ARMED ROBBERY
WITH KIDNAP, ARSON OR BURGLARY.
THE NEXT CATEGORY IS ARMED ROBBERY WITH MULTIPLE
VICTIMS.
NEXT IS ARMED ROBBERY WITH SOME OTHER CONTEMPORANEOUS
OFFENSE. |
AND YOU CAN SEE THAT THESE TEND TQ BE THE MORE
AGGRAVATED TYPE OF ARMED ROBBERY CASES AND THE RACIAL EFFECTS
PERSIST, ALTHOUGH ONE NOTES THAT AGAIN THE SAMPLE SIZES START TO
DIMINISH WHICH IS INEVITABLY THE CASE WHEN YOU BEGIN TO CONTROL
FOR A NUMBER OF FACTORS USING THIS KIND OF METHODOLOGY OF CROSS
asa
BALDUS ~ DIRECT
1 | TABULATION, YOU BEGIN TO LOSE SAMPLE SIZE.
rN
HOWEVER, I THINK IT’S IMPORTANT TO NOTE THAT THE KEY
|
3 | CHARACTERISTIC OF THESE TABLES IS THE PATTERN. THAT EVEN THOUGH
3 YOU LOSE STATISTICAL SIGNIFICANCE AS THE SAMPLE SIZE DROPS. YOU
(R
STILL SEE THE PATTERN OF THE EFFECT. IT”S NOT THE LACK OF
& EFFECT THAT LOSES THE SIGNIFICANCE, BUT IT’S THE LACK OF THE
? SAMPLE SIZE. THE DROP IN THE SAMFLE SIZE THAT AFFECTS THE LEVEL
8 OF STATISTICAL SIGNIFICANCE.
? I MAKE THAT COMMENT AS A GENERAL OBSERVATION ABOUT
0 INTERPRETING DATA WHEN ONE IS TRYING TO CONTROL FOR FACTORS. AND
11 IN THE PROCESS REDUCES THE SAMPLE SIZE. THE PURPOSE OF A TEST
12 OF STATISTICAL SIGNIFICANCE IS TO GIVE YOU A SENSE OF THE
12 LIKELIHOOD THAT WHAT YOU‘RE SEEING IS THE RESULT OF CHANCE
14 FACTORS. RATHER THAN REAL EFFECTS IN THE SYSTEM.
15 A TEST OF STATISTICAL SIGNIFICANCE SPEAKS TO ONE
1& SAMPLE. WHEN ONE SEES A PATTERN OF RESULTS ACROSS MANY SAMPLES
17 WITH THE SAME EFFECT DEMONSTRATED, THAT IS ADDITIONAL EVIDENCE
18 THAT SUGGESTS THAT IS NOT A CHANCE RESULT, EVEN THOUGH ALL WERE,
w 19 A NUMBER OF THEM MAY NOT THEMSELVES INDEPENDENTLY ACHIEVE
20 STATISTICAL SIGNIFICANCE.
21 FINALLY. WE GOT DOWN TO THE GROUP WHERE WE HAD ARMED
22 ROBBERY AND THERE WAS NO OTHER CONTEMPORANECUS OFFENSE. AND THAT
23 CONSTITUTES A FAIRLY LARGE POPULATION OF CASES. THOSE ARE THE
24 CASES SPECIFIED UNDER 2B SUBR4 ON THE BOTTOM OF THE FIRST PAGE OF
Fa] DB-846, AND THIS PROVIDED AN OPPORTUNITY TO FURTHER SUBDIVIDE THE
HD
~N
>
AB
0
8389
BALDUS -~ DIRECT
CASES, AND CONTROL ON THE BASIS OF QUALITATIVE JUDGMENTS ABOUT
WHAT WERE THE IMPORTANT FACTORS IN THESE CASES THAT WOULD
DETERMINE THEIR DEGREE OF DEATH WORTHINESS. AND WE ESTABLISHED
A SYSTEM FOR CATEGORIZING THESE CASES.
THE COURT: "MAY I ASK YOU A QUESTION --
THE WITNESS: YES, YOUR HONOR.
THE COURT: ~—- BEFORE YOU LEAVE WHAT YOU JUST TESTIFIED
TQ.
IF 1“M READING THIS CHART RIGHT, THE COLUMN, OR LINES
ROMAN NUMERAL II, CAPITAL A AND B THROUGH IV DO SHOW A PATTERN
OF DISCRIMINATION BASED ON THE RACE OF THE DEFENDANT, BUT DO NOT
ON BALANCE SHOW A PATTERN BASED ON RACE OF THE VICTIM.
IS THAT FAIR?
THE WITNESS: NO, YOUR HONOR. THE, THE WAY THIS TABLE
IS CONSTRUCTED IS AS FOLLOWS:
THE COLUMN C AND D, IF YOUR HONOR WOULD LOOK AT THE
‘CAPTION ON COLUMN C AND D, YOU CAN SEE THERE THE WAY WE
DESCRIBED THIS, THIS IS A COLUMN THAT MERSURES RACE OF VICTIM
EFFECTS CONTROLLING FOR THE RACE OF THE DEFENDANT. THAT IS. IN
THESE TWO COLUMNS, WE HAVE IDENTIFIED THE CASES THAT INVOLVE A
BLACK DEFENDANT AND A WHITE VICTIM AND A BLACK DEFENDANT AND A
BLACK VICTIM. BY DOING THAT WE SAY WE HAVE CONTROLLED ON,
BLOCKED ON THE RACE OF THE DEFENDANT.
THE THING THAT VARIES BETWEEN THESE TWO SUBSETS OF
CASES IS THE RACE OF THE VICTIM. THE RACE OF THE DEFENDANT IS
8460
BALDUS - DIRECT
1 (CONSTANT IN THE CASES, AS IS ALSO CONSTANT THE TYPE OF OFFENSE
AND THE OTHER CHARACTERISTICS THAT ARE SPECIFIED IN THE 2
3 LEFT-HAND MARGIN. SO THIS MEASURES, THIS TABLE, 38B, MEASURES
4 RACE OF VICTIM EFFECTS, AND YOU CAN SEE QVER ON THE RIGHTHAND
3 SIDE OF DB-8&. COLUMNS G AND H. THERE WE ARE CONTROLLING AGAIN
a & |FOR THE RACE OF THE DEFENDANT BY —-
~J
THE COURT: THATS WHAT I THOUGHT I WAS LOOKING AT.
© NOW. LET ME COME BACK AT YOU, LOOKING AT YOUR DATA
? ANOTHER WAY.
10 THE WITNESS: YES.
11 THE COURT: WE HAVE TWO RACIAL VARIABLES. ONE IS THE
12 DEFENDANT, AND ONE IS THE VICTIM.
13 IN A SOCIETY WHERE YOU MAKE THE ASSUMPTIONS THAT YOU
14 MAKE BASED ON THE LITERATURE, WHICH I BELIEVE IS THE NICE WAY
15 YOU HAVE PUT IT HERETOFORE, IF SOCIETY WERE REALLY SELECTING
16 KILLERS OF WHITES TO THE EXCLUSION OF BLACKS AND YOU CONTROLLED
17 FOR THE RACE OF THE DEFENDANT, THEN YOU SHOULD SEE THAT WHERE
18 THE DEFENDANT IS WHITE AND THE VICTIM IS WHITE AND BLACK. BUT
19 IF I’M READING THIS TABLE RIGHT, THE IMPOSITION OF THE DEATH
20 PENALTY RATE IS THE SAME. UNDER 5 AND H, WHERE YOU'VE GOT A
21 WHITE DEFENDANT.
22 SO WOULD IT BE FAIR TO SAY THAT THIS TABLE SAYS THAT
23 SOCIETY GETS UP TIGHT WHEN A BLACK DEFENDANT KILLS A WHITE
24 VICTIM, WHICH MAY BE WRONG, BUT ISN-T. ARE NOT WE SAYING
23 SOMETHING MORE THAN IT JUST GETS UPSET WHEN THE VICTIM IS WHITE?
BALDUS - DIRECT
THE WITNESS: WELL, YOU CAN SEE COVER HERE, YOUR HONOR,
WHEN YOU LIMIT THE INQUIRY TO WHITE DEFENDANT CASES, IN G AND H,
AND WE EXAMINE THE IMPACT OVER THERE OF THE RESPONSE OF THE
SYSTEM IN HANDLING WHITE VICTIM CASES AND BLACK VICTIM CASES,
WHEN THE DEFENDANT IS WHITE, --
THE COURT: THERE-S NO DISPARITY.
THE WITNESS: WELL, THE OVERALL EFFECT SHOWS A SLIGHT
DISPARITY. IT‘S MUCH LESS STABLE. YOU CAN SEE THE WHITE ON
WHITE CASES. YOUR HONOR, THE DEATH SENTENCING RATE IS SLIGHTLY
HIGHER. IF WE LOOK UP IN THE TOP COLUMN, TOP ROW, ROW ONE, AND
YOU AGGREGATE THE CASES. YOU CAN SEE THAT THERE IS A DIFFERENCE
IN THE, IN THE RATE, 12 PERCENTAGE POINTS. IT/S NOT AS LARGE TO
BE SURE.
THE OTHER POINT IS THAT IT CONSTITUTES A MUCH SMALLER
POPULATION IN THE SENSE THAT YOURE DEALING WITH RELATIVELY FEW
WHITE DEFENDANT-BLACK VICTIM CASES.
ALSO IN TERMS OF WHAT WE WOULD EXPECT, YOUR HONOR, THE
LITERATURE SUGGESTS THAT CASES THAT INVOLVE INTERRACIAL CRIMES
WOULD BE EXPECTED TO PRODUCE A SLIGHTLY MORE PUNITIVE RESPUNSE.,
REGARDLESS OF THE RACE, THE DEFENDANT-VICTIM RACIAL COMBINATION,
THATS ANOTHER PATTERN ONE SEES IN THE LITERATURE.
THE COURT: THAT IS SURPRISING TO ME.
THE WITNESS: PARDON ME?
THE COURT: THAT IS SURPRISING TO ME.
THE WITNESS: AND THAT COMBINED WITH THE FACT THAT
M
O
N
G
C
R
h
y
N
e
po
t QD
f
o
o
e
BALDUS - DIRECT
YOURE DEALING WITH A VERY SMALL PROPORTION OF WHITE ON BLACK
CASES HERE, YOUR HONOR, THE, AND I THINK IT’S THE COMBINATION OF
THOSE TWO EFFECTS THAT EXPLAINS WHY YOU FIND INSTABILITY IN
THESE EFFECTS. |
BUT, THERE CERTAINLY DOES NOT SUGGEST THAT THE BLACK
VICTIM CASES ARE TREATED MORE PUNITIVELY, WHEN THE, THE VICTIM,
|PARDON ME, WHEN THE DEFENDANT IS WHITE. I MEAN THATS CERTAINLY
NOT WHAT G AND H ARE TELLING ME.
THE CQURT: I“M TRYING TO USE WHAT YOU“RE TELLING ME TO
USE, AND THAT 1S THE PERSISTENCY OF THE PATTERN.
THE WITNESS: YES.
THE COURT: AND IM LOOKING DOWN COLUMNS I AND J, YOUR
PROPOSITION IS SUPPORTED IN THE FIRST LINE: IT IS NOT IN THE
SECONDS IT IS NOT IN THE THIRD: IT IS IN THE FOURTH: IT IS NOT
IN THE FIFTH; AND IT IS IN THE SIXTH.
THE WITNESS: THAT'S RIGHT.
THE COURT: AND THERE AIN’T NO PATTERN THERE.
THE WITNESS: THAT’S RIGHT. THOSE RESULTS ARE MIXED.
NO, I QUITE AGREE WERE YOU. BUT YOQU-RE DEALING WITH A VERY
SMALL POPULATION.
THE COURT: WELL, THAT WOULD BE A STANDARD OBJECTION
ANYTIME YOU DO IT. BUT WHAT YOU'RE TRYING TO TELL ME IS THAT
I CAN LOOK AT FERSISTENCY THROUGHOUT A NUMBER OF STATISTICALLY
INSIGNIFICANT, OR ANALYSIS WHICH, EACH ONE OF WHICH WOULD BE IN
AND OF THEMSELVES STATISTICALLY INSIGNIFICANT. IF I SEE AR
8463
BALDUS ~ DIRECT
PATTERN PERSIST, THEN I MAY CREDIT THE FACT THAT THAT PATTERN IS
THERE. AND WHAT I-M TELLING YOU, THAT DOESN’T SHOW THAT.
IF THE ANALYTICAL FRAMEWORK YOU HAVE GIVEN ME IS TRUE,
YOUR CONCLUSION IS WRONG AS TO THE DATA ON THIS FIRST PAGE, I
THINK. NOW, YOU TELL ME WHY IT ISN“T.
THE WITNESS: I WILL TRY TO, YOUR HONOR.
AS 1 SUGGESTED YESTERDAY, THE PRINCIPAL SOURCE OF THE
RACE OF THE VICTIM DISPARITIES IN THE SYSTEM IS THE BLACK
DEFENDANT CASES. THAT’S THE PRINCIPAL SOURCE.
THE WAY THOSE CASES ARE TREATED, THEY CONSTITUTE THE
BIGGEST BULK OF THE CASES IN THE SYSTEM. IT’S THE WAY THOSE
CASES ARE TREATED AS BETWEEN THE WHITE VICTIM AND BLACK VICTIM
CASES THAT IS THE PRINCIPAL SOURCE OF THE OVERALL DISPARITIES
THAT WE SEE IN THE, AVERAGE DISPARITIES THAT WE“RE MEASURING IN
THE RACIAL CO-EFFICIENTS. I THINK THIS --
THE COURT: SAY THAT AGAIN, THE PRINCIPAL SOURCE -—
THE WITNESS: THE PRINCIPAL SOURCE OF THE RACE OF
VICTIM DISPARITY.
THE COURT: IS IN COLUMN B, IN ESSENCE. THE WITNESS: YES. IT’S THE MEASURE YOU SEE IN COLUMN E
AND F. YOUR HONOR.
YOU“LL SEE HERE THAT THESE DISPARITIES ARE
SUBSTANTIALLY LARGER THAN ANY DISPARITIES WE SEE REFLECTED IN AN
OVERALL REGRESSION ANALYSIS, AND WHAT YOURE OBSERVING IS THAT
INDEED AS YOU POINT OUT THE EFFECTS ARE NOT CONSISTENT ACROSS
8464
BALDUS - DIRECT
1 |EACH SUB-GROUP OF CASES WHEN YOURE TALKING ABOUT WHITE
DEFENDANT CASES.
THATS VERY PLAIN AND IT’S PLAIN IN EVERY TYPE OF
ANALYSIS WE DO.
HOWEVER, WHEN YOU TAKE AN AVERAGE OF ALL THESE EFFECTS.
o
e
F
S
ER
S
E
YOU CONTINUE TO SEE A STRONG PERSISTENT OVERALL EFFECT.
7 WHEN YOU LOOK AT THE SUB-GROUP OF CASES. WHICH PROVIDE
a THE SOURCE OF THESE DISPARITIES, THAT IS, THE BLACK DEFENDANT,
9 THERE YOU SEE MUCH LARGER EFFECTS, AND LARGER AND PERSISTENT.
10 AND USING THE RULE THAT I WAS SUGGESTING TO YOUR HONOR,
11 IF YOU LOOK OVER THE ENTIRE TABLE, 38B, YOU LOOK DOWN COLUMNS E
12 AND F IN THE ENTIRE TABLE, YOU FIND ONLY ONE PLACE THAT THE
13 PATTERN DOES NOT EXIST, OR TWO PLACES. AT ONE SPOT WHERE THE
14 DEFENDANT IS NOT THE ACTUAL KILLER, ON THE BOTTOM OF PAGE 2 OF
15 DB-B6, YOU SEE THE RACIAL EFFECTS DISAPPEAR. AND SIMILARLY» ON
16 THE PAGE 3 OF THIS, THE RACIAL EFFECTS DISAPPEAR IN A CLASS OF
17 CASES WHERE THERE ARE NO DEATH SENTENCES.
18 THIS 1S PERFECTLY CONSISTENT WITH THE PATTERN, IN THE
@ 19 LOW AGGRAVATED CASES. YOU DON’T SEE RACIAL EFFECTS. IT’S ONLY IN
20 THE MORE AGGRAVATED CASES AND IN THOSE CASES, IT’S PRINCIPALLY A
21 REFLECTION OF THE DISPARITY IN TREATMENT OF BLACK DEFENDANTS
22 WHEN THE VICTIM IS WHITE AND WHEN THE VICTIM IS BLACK.
23 | THE COURT: 1 HEAR YOU AND I AM WONDERING IF WE'RE NOT
z4 AGREEING ON THE BASIC PROBLEM. BUT DISAGREEING ON HOW THAT IS
<3 CHARACTERIZED.
e
T
T
e
El
Po
o
N.
f
o
12
13
2635
BALDUS - DIRECT
IT WOLD BE FAIR TO SAY FROM LOOKING AT COLUMN C AND D
THAT SOCIETY IS NOT STRONGLY SANCTIONING THE CRIME OF A BLACK
KILLING A BLACK, AND IT IS IN MANY CASES MORE STRONGLY
SANCTIONING A BLACK KILLING A WHITE. AND THAT PATTERN SEEMS TO
BE PERSISTENT.
BUT IF YOU BREAK THEM DOWN LIKE YQU-“VE DONE IT HERE, I
BEGIN TO WONDER ABOUT THE OVERALL PREMISE THAT IT IS THE RACE OF
THE VICTIM THATS CONTROLLING.
IN ANY EVENT, I HAVE YOUR POINT. I HAVE THAT QUESTION
WHICH YO MIGHT FIND SOME WAY TO ADDRESS IN ANOTHER AREA.
BY MR. BOGER:
?. PROFESSOR BALDUS. FOR CLARITY. WHAT WAS THE PURPOSE FOR
CONDUCTING THIS BREAKDOWN IN DB-847
A. OH, THIS IS, YOU’RE REFERRING, COUNSEL, TO PART 2, SUB-B.,
SUB 47
@. NO, I WAS IN GENERAL ASKING WHY YOU WERE UNDERTAKING THIS
KIND OF ANALYSIS. WHAT WERE YOUR OBJECTIVES?
THE COURT: I THINK I KNOW WHERE HE WAS GOING, LET ME
MAKE SURE I PICKED UP ON IT, BECAUSE IT WAS INTERESTING. WITH
WHAT YOU SAID THE OTHER DAY. AND THAT IS, THAT DISPARITY
DECREASES AS THE AGGRAVATING OFFENSES DECREASE, AND IS THAT ONE
OF THE THINGS YOU WERE LOOKING FOR HERE?
THE WITNESS: YES, SIR.
THE COURT: ALL RIGHT. G0 AHEAD WITH THAT IDEA, THEN.
MR. BOGER: FINE.
8&b6
BALDUS - DIRECT
THE WITNESS: YES, YOUR HONOR.
THE PURPOSE OF THIS ANALYSIS WAS TO TEST THIS GENERAL
HYPOTHESIS ABOUT THE RELATIONSHIP BETWEEN THE LEVEL OF RACIAL
DISPARITY OBSERVED AND THE SERIOUSNESS OF THE CASES. THAT'S
BEEN A COMMON THEME THAT WE“VE EXPLORED.
IN ADDITION, HERE THE OBJECTIVE WAS TO USE AN
ALTERNATIVE METHODOLOGY TO EXPLORE THE QUESTIONS THAT WE HAVE
HERETOFORE EXPLORED STRICTLY WITH STATISTICAL METHODOLOGY
BECAUSE HERE WERE ACTUALLY LOOKING AT CASES, THAT IS, SUMMARIES
OF CASES, AND MATCHING THEM, PAIRING THEM TOGETHER. ALL THESE
PAIRINGS WERE MADE WITHOUT ANY KNOWLEDGE OF THE RACE OF THE
VICTIM. AND WITHOUT ANY KNOWLEDGE OF THE SENTENCE, THE PEOPLE
THAT WERE INSTRUCTED TO DQ THIS WERE ASKED TO DO THIS IN A
COMPLETELY BLIND FASHION WITH RESPECT TO THESE VARIABLES. AND --
BY MR. BOGER:
RN. WE’RE TALKING NOW ABOUT THE SORTING ON PAGE 2 OF DB-387
A. THAT'S RIGHT. IN TERMS OF TRYING TO EXAMINE THESE CASES,
AND DETERMINE WHAT THE FACTORS WERE THAT WOULD BE IMPORTANT IN
SEPARATING THE MORE AND LESS AGGRAVATED CASES.
WHAT WE CAME UP WITH. YOUR HONOR. AND THIS ENABLES US TO
FURTHER REFINE THE DEFINITION OF A POOL OF COMPARABLE OR SIMILAR
CASES. IT RECOGNIZES THAT THERE ARE LIMITS ON WHAT STATISTICAL
ANALYSIS CAN DO IN THIS AREA AND WE WANTED TO SEE IF WE DID THIS
IN A MORE-FINE GRAIN FASHION. IN A MANNER MORE COMPARABLE TO
QUALITATIVE TYPE OF CASE MATCHING. WHETHER WE WOULD SEE
a
hb
J
1
B
O
BE
8467
BALDUS - DIRECT
| COMPARABLE RESULTS, AND WE ESTABLISHED A SET OF CRITERIA TO
BREAK THE CASES INTO 3 CATEGORIES. WE USED 3 CATEGORIES
BASICALLY BECAUSE IF WE SLICED IT UP MUCH MORE FINELY. WE WOULD
LOSE ALL SAMPLE SIZE AND THESE SEEMED INTUITIVELY SENSIBLE AND
| COMPREHENSIBLE IN TERMS OF, AT LEAST MY JUDGMENT, TO THE OVERALL
SERIOUSNESS OF THESE CARES.
fl. WHAT WERE THE STANDARDS OR MEASURES OF LESS AGGRAVATED ARMED
ROBBERY CASES. TYPICAL ARMED ROBBERY CASES, AND MORE AGGRAVATED
ARMED ROBBERY CASES THAT YOU EMPLOYED?
A. YES. THE, TO JUST GIVE AN OVERVIEW, WE'RE STARTING NOW WITH
ARMED ROBBERY CASES. WAS THE FIRST VARIABLE WE WERE SORTING ON,
AND WE RESTRICTED TO A CLASS WHERE THERE WAS NO OTHER
CONTEMPORANEOUS OFFENSE.
| WE THEN FURTHER SORTED THOSE CASES INTO A CATEGORY OF
MORE AGGRAVATION, MORE TYPICAL AND LESS, WERE THE THREE
CATEGORIES.
THE MORE AGGRAVATED CASES CONSISTED OF THOSE THAT HAD
ONE OR MORE oF THE FOLLOWING.
MULTIPLE VICTIMS, PRIOR HISTORY OF ARMED ROBBERY OR
HOMICIDE. CLEAR PREMEDITATION TO MURDER, GRATUITOUS INFLICTION
OF PAIN, MUTILATION, HELPLESS VICTIM. THE DEFENDANT ON ESCAPE OR
ON PAROLE. THAT’S WHAT RESULTED IN THE CLASSIFICATION OF A CASE
IN THE MORE AGGRAVATED CATEGORY.
THEN WE IDENTIFIED THE MITIGATING FACTORS THAT SEEM TO
JUSTIFY DISTINGUISHING CASES FROM THE TYPICAL, RUN OF THE MINE
BALDUS ~ DIRECT
r3 3 CASE, AND THOSE WERE CASES IN WHICH THE DEFENDANT WAS NOT THE
TRIG0ER MAN, THERE WAS NO INTENT TO KILL ON THE PART OF THE
DEFENDANT, THERE WAS EVIDENCE THAT THE DEFENDANT ACTED TO DEFEND
HIMSELF IN SOME WAY ALTHOUGH IT WAS NOT SUFFICIENT TO OBVIOUSLY
SUSTAIN A THEORY OF SELF-DEFENSE. DEFENDANT HAD NO PRIOR FELONY
CONVICTIONS, DEFENDANT WAS UNDER THE INFLUENCE CLEARLY OF
ALCOHOL OR DRUGS, WITH A, THE DEFENDANT WAS MENTALLY DEFICIENT.
0
~~
A
a
bb
W
N
| WE SORTED THOSE CASES TOGETHER INTO A CATEGORY THAT WE
? CALL LESS AGGRAVATED.
10 THE COURT: I DON’T HAVE THAT. HOW MANY PAGES IS THIS
11 EXHIBIT SUPPOSED TO BE?
$2 THE WITNESS: OH, IT’S NOT EXHIBIT. YOUR HONOR. I HAVE
13 A COPY OF IT HERE. |
14 THE COURT: GO AHEAD.
13 BY MR. BOGER:
16 R. I THINK IT’LL BE PART OF YOUR FINAL REPORT. FROFESSOR
17 BALDUS?
18 A. YES.
w 19 @. DID YOU DO THE SORTING OF CASES, A PRIORI AND THEN DEVELOP
20 THE CATEGORIES THAT YOU‘VE JUST DEFINED AND SEPARATE THEM. OR
21 DID YOU DEVELOP THE CATEGORIES AND THEN SORT THE CASES ON THE
22 BASIS OF THESE CRITERIA?
23 A. WELL, IT’S AN INTERACTIVE PROCESS. YOU START WITH A SET OF
24 CATEGORIES THAT YOU THINK WILL SORT THESE CASES. CATEGORY THAT
25 WE DEVELOPED IN OTHER CASES, BUT WE FOUND WHEN WE STARTED LOOKING
|
i
i
BALDUS ~ DIRECT
AT THESE CASES THAT THEY HAD SOME NUANCES THAT THE OTHER CASES
THAT WE HAD LOOKED AT USING THIS SAME PROCEDURE DID NOT.
SO YOU MODIFY THEM. BUT THE OBJECT IS TO TRY AND
IDENTIFY THE CHARACTERISTICS OF THE CASES THAT PROVIDE SOME
MEASURE OF OVERALL CULPABILITY AND WE BELIEVE THEY WOULD BE
PERCEIVED IN THE SYSTEM.
@. WHAT TOOL DID YOU USE TO EXAMINE THE CASES THEMSELVES?
A. THE SUMMARY OF THE CASES. THAT IS, WE HAD A TYPED COPY OF
EACH OF THE SUMMARIES OF THESE CASES THAT WERE PROVIDED BY THE
CODERS AS A RESULT OF THEIR RESEARCH.
2. AND WHAT DO THOSE RESULTS SHOW?
A. WELL, THEY SHOW FIRST OF ALL THAT OUR JUDGMENTS ABOUT THE
CASES WHICH ARE MORE SERIQUS IS REFLECTED BY THE ACTUAL PATTERN
OF SENTENCING DECISIONS IN THE SYSTEM, THAT IS, THE CASES IN THE
LESS AGGRAVATED CATEGORY, AS IS REFLECTED IN COLUMN B OF PAGE 2
OF DB-84, SHOWS A 13 PERCENTAGE RATE, 13 PERCENT DEATH
SENTENCING RATE, THAT’S SIX OUT OF FORTY-FIVE OFFENDERS IN THAT
CATEGORY.
THE COURT? WHAT LINE ARE YOU ON?
THE WITNESS: THIS IS PAGE 2, YOUR HONGR, LINE 4, IT’S
CALLED “LESS AGGRAVATED ARMED ROBRERY WITH NO OTHER OFFENSE."
THE COURT: OKAY.
THE WITNES3: AND WE LOOK IN THERE AMONG THE, AT THE
AVERAGE DEATH PENALTY CASE, RATE, RATHER, AMONG THOSE CASES, AND
WE SEE THAT SIX OF THOSE OFFENDERS RECEIVED A DEATH SENTENCE OUT
— — ar areas S————
MN
Gi
870
BALDUS - DIRECT
OF FORTY-FIVE.
AND THEN WE LOOK AT THE TYPICAL CATEGORY AND WE SEE A
SLIGHT ELEVATION IN TERMS OF THE OVERALL DEATH SENTENCING RATE
FICKS UP NOW TO 18.
AND THEN WE LOOK IN THE MORE AGGRAVATED CATEGORY. THOSE
CASES HAVING THE CHARACTERISTICS THAT 1 DEFINED, PUT THEM IN
THAT CATEGORY AND WE SEE A SHARP INCREASE IN THE DEATH
| SENTENCING RATE AMONG THOSE CATEGORIES.
THEN IF YOU LOOK OVER, YQUR HONOR, INTO, UNDER COLUMN
E, YOU SEE THE SAME PATTERN THAT 1 POINTED OUT IN OTHER
OCCASIONS. THAT IS, AS THE DEATH SENTENCING RATE RISES. SO DOES
THE RACE OF VICTIM DISPARITY AMONG THIS CLASS OF CASES, AMONG
BLACK DEFENDANT CASES.
AND THAT 5 A MAIN THEME OF THE, OF THIS. THATS A MAIN
PATTERN THAT EMERGES FROM THIS, AND ALL OF THE ANALYSIS THAT WE
HAVE DONE, AND THE IDEA HERE IS TO SEE WHETHER OR NOT THIS SORT
OF PATTERN EMERGES WHEN WE USE THIS NON-STATISTICAL QUALITATIVE
METHOD OF SORTING CASES.
AND WE SEE FINALLY IN THIS CATEGORY WHERE THE DEFENDANT
WAS NOT THE ACTUAL TRIGGER MAN. THAT ACTUALLY: THIS IS SLIGHTLY
MISLEADING HERE, YOUR HONOR. THAT IS NOT ONE OF THE
SUB-CATEGORIES HERE. THAT'S A WHOLE DIFFERENT CATEGORY WHERE
THE DEFENDANT IS NOT THE TRIGGER, WE SEE THE DEATH SENTENCE RATE
1S VERY LOW AND THE RACE OF VICTIM DISPARITIES IN THIS CATEGORY
DISAPPEAR.
871
pALDUS ~ DIRECT
pb
AND THEN FINALLY. ON 43, WE LOOK AT THE OTHER BZ
CONTEMPORANEOUS OFFENSES. AND HERE WE LIMITED THE ANALYSIS
SIMPLY TO CONTROL FOR THOSE SEPARATE FACTORS. THAT IS. THE
PRESENCE OF THOSE FACTORS IN THE CASE.
AND AGAIN, YOU SEE A PATTERN THAT IS CONSISTENT WITH
WHAT THE PATTERN WAS IN THE EARLIER ANALYSIS.
BY MR. BOGER:
@. PROFESSOR BALDUS, DO YOUR EXPERT OPINIONS RELY SOLELY ON
vw
0
~N
Q
o
s
O
N
THIS MODE OF ANALYSIS?
10 A. NO. THIS IS JUST ONE OF A VARIETY OF DIFFERENT ANALYSES.
11 AND THEY CONFIRM WHAT THE STATISTICAL ANALYSIS SHOWS. AND
12 THATS WHAT GIVES ME CONFIDENCE IN THE RESULT. YOUR HONOR. IS
13 THAT. THAT THE ANALYSIS WITH RESPECT TO THE RACE OF THE VICTIM
14 PRODUCES COMPARABLE RESULTS. NO MATTER HOW WE SLICE THE DATA.
13 IF I WERE TO LOOK AT THESE WITHOUT DOING A STATISTICAL
16 ANALYSIS, I WOULDN‘T HAVE, I HAVE QUESTIONS. IF I DID THE TIME
17 STATISTICAL ANALYSIS WITHOUT LOCKING AT THE QUALITATIVE. 1
13 WOULD. IF I DID ONE STATISTICAL ANALYSIS AND IT SHOWED ONE
w 19 THING AND ANOTHER, ANOTHER RESULT, I WOULD HAVE QUESTIONS ABOUT
20 THERE BEING ANY SYSTEMATIC EFFECT IN THE SYSTEM. ITS THE
|
|
|
{
PATTERN THAT TELLS THE STORY, IN MY OPINION.
22 MR. POGER: YOUR HONOR, AT THIS TIME I MOVE THE
23 ADMISSION OF DB-8& INTO EVIDENCE.
24 MS. WESTMORELAND: YOUR HONOR, AT THIS POINT ON THIS
23 EXHIBIT, I STATE ALL THE OBJECTIONS I-VE HAD FOR THE PRECEDING
Oo
~ br
BALDUS - DIRECT
Pa
. EXHIBITS.
THE COURT: I WILL ADMIT IT, NOTING THIS FOR COUNSEL.
THAT I HAVE LESS CONFIDENCE IN THE PROBATIVE VALUE OF THIS
EXHIBIT THAN I DO SOME OF THE OTHERS THAT I°VE ACTUALLY ADMITTED
IN EVIDENCE. TWO REASONS: IM NOT ENTIRELY SURE OF THE
CRITERIA. ON PAGE 4 I HEARD WHAT IT SAID, BUT I“M NOT SURE
WHAT 1S MORE OR LESS AGGRAVATING, AND ALSO NOTING THAT THERE IS
NO EFFORT MADE TO CONTROL FOR SOME VARIABLES WHICH I THINK I
D
N
e
h
C
W
HAVE LEARNED ARE ALMOST ALWAYS GOING TQ OBSCURE THE RESULTS.
pt
<>
BUT IT“S ADMITTED.
MR. BOGER: THANK YOU, YOUR HONOR. PS
N
fo
wy
i2 BY MR. BOGER:
13 @. PROFESSOR. LET ME ASK YOU TO TURN TQ DB-87 MARKED FDR
14 IDENTIFICATION. WHAT DOES THIS DOCUMENT DO?
15 A. THIS DOCUMENT IS A PRESENTATION OF THE RACE OF THE DEFENDANT
146 DISPARITIES IN THE SAME CATEGORIES OF CASES WE JUST LOOKED AT IN
17 DB-26.
18 : VERY SIMPLY, WHAT WE DID WAS REARRANGE THE COLUMNS
Ww 19 CONCERNING THE DEFENDANT-VICTIM RACIAL COMBINATIONS SO THAT. FOR |
20 EXAMPLE, IN COLUMNS C AND D, WE’RE NOW CONTROLLING FOR THE RACE
21 OF THE VICTIM. AND DEMONSTRATING THE DIFFERENTIAL TREATMENT IN
22 THE RACE OF THE DEFENDANT.
>3 AND HERE OUR PARTICULAR FOCUS WAS ON THE TREATMENT OF
<4 BLACK OFFENDERS IM THE CONTEXT OF A WHITE VICTIM CASE, BECAUSE
23 THATS WHERE ALL OF OUR EVIDENCE UP TO THIS POINT HAD SUGGESTED
{
|
BALDUS - DIRECT
WAS THE SOURCE OF ANY OF THE RACE OF DEFENDANT DISPARITIES THAT
WE MIGHT OBSERVE IN THE SYSTEM.
AND WHEN YOU EXAMINE THIS ANALYSIS YOU) SEE THE SAME
PATTERN, YOU SEE DISPARITIES BASED UPON THE RACE OF THE
DEFENDANT IN THIS CATEGORY OF CASES. BUT THEY AREN‘T NEARLY AS
STRONG AS RACE OF VICTIM EFFECTS, AND THERE-S A CERTAIN AMOUNT
OF INSTABILITY IN THOSE RESULTS.
WE LOOK DOWN COLUMNS E AND F, ACROSS THESE 3 PAGES, YOU
SEE SOME SITUATIONS WHERE THE RATE FOR THE BLACK DEFENDANT IS,
WITH A WHITE VICTIM IS LOWER THAN IT IS WITH A WHITE DEFENDANT
WITH A WHITE VICTIM.
BUT OVERALL YOU GET A SENSE THAT THERE IS A PATTERN
REFLECTING A CERTAIN LEVEL OF DISADVANTAGE FOR A BLACK DEFENDANT
WITH A WHITE VICTIM VIS-A-VIS A WHITE DEFENDANT WITH A WHITE
VICTIM.
THE COURT: THIS PROBABLY HAS NOTHING TO DO WITH THE
LAWSUIT, BUT 1 WOULD OBSERVE THAT THIS TABLE COMES CLOSER TO
PREDICTING SOUTHERN LORE THAN ANY IVE SEEN IN ANY OF THE
STUDIES. THE LORE WOULD BE THAT THE SOUTH DOES NOT TOLERATE
BLACK AGGRESSIVE ACTS AGAINST WHITES. AND THAT OTHER THINGS
BEING EQUAL AND THAT NOT BEING THE SITUATION, IT HOLDS WHITE
DEFENDANTS TO A HIGHER STANDARD THAN IT HOLDS BLACK DEFENDANTS.
AND I THINK THAT TABLE MAY DEMONSTRATE SOME OF THAT.
| LOOKING AT THE FIRST PAGE. WHERE YOU HOLD THE RACE OF
THE VICTIM CONSTANT, IF I READ THIS TABLE RIGHT, THE SYSTEM
nm
I
|
| BALDUS - DIRECT
|
RESPONSE IS TO HOLD THE WHITE DEFENDANT MORE CULPABLE OR AT
LEAST TO IMPOSE A STRONGER PENALTY THAN AGAINST THE BLACK
DEFENDANT. IS THAT RIGHT?
THE WITNESS: THE COMPARISON OF C AND D, YOUR HONOR,
YOU“RE TALKING ABOUT, IN THE --
THE COURT: NO. IM LOOKING AT G AND H, WHERE WE HAVE A
BLACK VICTIM AND WE’RE STUDYING HOW THE SYSTEM RESFONDS TO THE
KILLING OF A BLACK VICTIM, IT WOULD, WOULD IT NOT. BE FAIR TO
SAY, BASED AT LEAST ON PAGE 1, AND I GUESS TO SOME EXTENT ON THE
OTHER PAGES. THAT WHERE THE RACE OF THE VICTIM IS BLACK, AND THAT
DOESN’T BRING UP ANY STEREOTYPES OR THAT SORT OF THING, BUT THE
SYSTEM IS HOLDING DR IMPOSING A GREATER PUNISHMENT ON THE
WHITE DEFENDANT THAN ON THE BLACK DEFENDANT RATHER REGULARLY
DOWN THROUGH PAGE ONE.
THE WITNESS: YES.
THE COURT: AND TO SOME EXTENT ON PAGE 2. AND MIXED BUT
NOT GUITE SO TRUE ON PAGE 3.
THE WITNESS: WELL, IF THAT IS. THAT. AS YOU DEFINE IT
AS SOUTHERN LORE. IN THE LITERATURE, THAT S RECOGNIZED AS
CROSSING LINES, CROSSING CAST LINES. CROSSING RACE LINES. THE
PEOFLE WHO DO CROSS LINES ARE GOING TO BE SANCTIONED MORE
HEAVILY. THAT THEORY WOULD TEND TO RECONCILE THE COMPARISON OF
G AND H. THAT IS, THE INTERRACIAL, INTERRACIAL CRIME IS GOING TO
BE TREATED MORE HARSHLY THAN INTRARACIAL CRIME. AND YOU CAN SEE
THAT BY COMPARING C AND D. IN "D" YOU HAVE AN INTERRACIAL: IN
BALDUS - DIRECT
| “D AN INTRARACIAL. AND IN H AND G, YOU HAVE THE SAME PATTERN,
INTER- AND INTRARACIAL. AND IN EACH CASE. IT’S THE INTERRACIAL
CRIME THAT SHOWS A SLIGHTLY MORE PUNITIVE RESPONSE, AND THAT IS
CONSISTENT WITH THE GENERAL SOCIOLOGICAL THEORY ABOUT THE
BEHAVIOR OF GROUPS.
THE COURT: I WON‘T BURDEN THE RECORD ANY FURTHER. BUT
I THINK YOU“VE GOT PART OF THAT WRONG. AS A SOUTHERN BOY. of
BE GLAD TO TELL YOU OVER A CUP OF TEA AFTER THIS IS ALL OVER.
MR. BOGER: YOUR HONOR, I MOVE THE ADMISSION OF DB-87
INTO EVIDENCE.
THE COURT: WITH THE NOTATIONS THAT I MADE ON THE
PREVIOUS EXHIBIT, IT-LL BE ADMITTED.
MR. BOGER: THANK YQU
YOUR HONOR, AT THIS TIME, I HAD INTENDED TO HAVE
PROFESSOR BALDUS IDENTIFY SOME NOTEBOOKS FROM WHICH HE DID
THESE TWO STUDIES. WE“VE JUST MENTIONED THAT ONE OF THOSE
NOTEBOOKS IS NOT WITH US THIS MORNING. I MAY WELL BRING THAT IN
THIS AFTERNOON, BUT SIMPLY WANTED TO ALERT YOU WHY WERE MOVING
OVER BEYOND DB-38.
THE COURT: ALL RIGHT.
BY MR. BOGER:
A. PROFESSOR BALDUS, HAVE YOU DONE ANY OTHER ANALYSIS, HAVE YOU
THOUGHT OF ANY OTHER METHOD THAT MIGHT PERMIT YOU TO DISTINGUISH
CASES BASED UPON THE RISK THAT THOSE CLASSES OF CASES MIGHT
RECEIVE A DEATH SENTENCE, IN ORDER TO DETERMINE WHAT THE RACIAL
BALDUS - DIRECT
1 EFFECTS ARE?
2 1A. YES.
3 RA. WHAT KIND OF ANALYSIS?
4 A. THE NEXT ANALYSIS THAT I UNDERTOOK WAS TO IDENTIFY THOSE
CASES IN THE ENTIRE SYSTEM THAT APPEARED TO BE AT GREATEST RISK ou
WITH THE, WITH RESPECT TO THE POSSIBILITY OF RECEIVING A DEATH
-w
Bb
SENTENCE. OUR CROSS TABULATION ANALYSES EARLIER HAD INDICATED
THAT WHEN THE LIKELIHOOD OF A DEATH SENTENCE IS VERY VERY LOW
0
©
BECAUSE THE CASE IS SIMPLY NOT AGGRAVATED AT ALL, THAT YOU
10 CANNOT REALLY CET A PICTURE OF HOW DISCRETION IS BEING
11 EXERCISED.
12 AND WE DEVELOPED A METHOD THAT WOULD ENABLE US TO
13 COMBINE A SUBSTANTIAL NUMBER OF BACKGROUND VARIABLES INTO AN
14 ANALYSIS IN WHICH WE WERE ABLE TO CONTROL FOR MANY BACKGROUND
13 FACTORS, SIMULTANEOUSLY. AND ASSESS THE SERIOUSNESS OF THE CASES
16 CONTROLLING FOR THESE BACKGROUND FACTORS.
17 AND THE BASIC PROCEDURE IS TQ DEVELOP A REGRESSION
1a ANALYSIS, MULTIPLE REGRESSION ANALYSIS, AND TD TAKE THE FACTORS |
w 19 THAT THE MULTIPLE REGRESSION ANALYSIS SUGGESTS ARE IMPORTANT
=0 PREDICTORS OF WHO RECEIVES A DEATH SENTENCE; THEN TO SCORE THE
ire) CASES IN TERMS OF THE PRESENCE OR ABSENCE OF THOSE FACTORS THAT
22 TEND TO MAKE A CASE MORE AGGRAVATED OR MORE MITIGATED. AND TO
23 THEN SORT THE CASES ACCORDING TO THE LEVEL OF AGGRAVATION OR
24 MITIGATION OR TO PUT IT MORE PRECISELY. THE PREDICTED LEVEL OF A
23 DEATH SENTENCE THAT ONE WOULD EXPECT, QRIVEN THE CHARACTERISTICS
k
s
| 877
BALDUS ~ DIRECT
OF EACH CASE.
SEE, ONCE ONE DOES ONE OF THESE ANALYSES WHERE THE
REGRESSION FROCEDURE INFORMS YOU OF WHAT THE SYSTEM SEEMS TO BE
RESPONDING TO, YOU CAN LOOK AT ONE INDIVIDUAL CASE AND SAY,
GIVEN OUR EXPERIENCE THROUGH ALL OF THESE. HOW LIKELY IS IT THAT
YOU THINK THIS PARTICULAR CASE WOULD RECEIVE A DEATH SENTENCE.
WE THEN TAKE THOSE CASES AND, AND RANK CRDER THEM
ACCORDING TO THIS PREDICTED LIKELIHOOD THEY WOULD RECEIVE A
DEATH SENTENCE, AND THEN BREAK THEM DOWN INTO SUB-GROUPS. AND
THE RESULTS THAT WE HAVE —-
@. BEFORE YOU REACH THE RESULTS, LET ME ASK YOU A QUESTION
ABOUT THE METHOD.
IS THERE ANY PRECEDENT FOR THAT KIND OF APPROACH TO
|THIS SORT OF QUESTION?
A. YES. THE, THE PRECEDENT THAT WE RELIED ON IN USING THIS IS Th
HALOTHANE STUDY THAT I MENTIONED IN MY EARLIER TESTIMONY.
AND YOUR HONOR WILL RECALL THAT WAS THE, THE STUDY THAT
WAS DONE TO TRY AND DETERMINE WHETHER HALOTHANE WAS DANGEROUS.
AND THEY HAD THE SAME PROBLEM THAT WE DID HERE IN THE SENSE THAT
THEY WERE TRYING TO DETERMINE HOW FREQUENTLY A FAIRLY RARE EVENT
OCCURS, WHILE CONTROLLING FOR A VARIETY OF BACKGROUND FACTORS
THAT BEAR UPON THE LIKELIHOOD THAT THAT WILL OCCUR, THAT IS, THE
CONDITION OF THE PATIENT, TYPE OF OPERATION, ET CETERA. ALL
iE
THOSE BACKGROUND FACTORS.
50, THE PROCEDURE THEY USED WAS NOT IDENTICAL TO THIS,
oO
N
B
A
o
H
W
N
873
BALDUS - DIRECT
BUT IT WAS COMPARABLE CONCEPTUALLY TO THIS, WOULD BE TO DEVELOP
Hii INDEX THAT WOULD TELL YOU HOW LIKELY A PERSON WOULD DIE FROM
OTHER FACTORS, AND TO SORT THOSE CASES INTO COMPARABLE
CATEGORIES. AND THEN YOU LOOK AT THOSE THAT HAD HALOTHANE AND
THOSE THAT USED SOME OTHER ANESTHETIC AND GET SOME SENSE OF WHAT
EFFECT THAT HALOTHANE WAS LIKELY HAVING IN THE PROCESS.
WE DID THE SAME THING HERE. WE LOOKED AT THE
BACKGROUND FACTORS THAT HAD A STRONG PREDICTIVE EFFECT IN THE
SYSTEM AND THEN SORTED THE CASES ACCORDING TO THOSE WHERE THE
PREDICTED LIKELIHOOD OF A DEATH SENTENCE WAS COMPARABLE, GIVEN
ALL THE LEGITIMATE NON-RACIAL BACKGROUND FACTORS. AND WE COULD
LOOK AT THE DIFFERENT RACIAL DISPARITIES WITHIN THESE
SUB~GROUPS.
| TO GIVE YOUR HONOR A SENSE OF WHAT THIS IS ANALOGUS TO,
I WOULD LIKE TO REFER BACK TO AN EARLIER FIGURE, IF I COULD.
YOUR HONOR, WHICH IS IN DB-74&. IF YOU’LL NOTE ALONG THE ROTTONM
OF DB-76 WE HAVE FOUR CATEGORIES OF CASES. AND THOSE INDICATE
THE DEATH SENTENCING RATES AFTER CONTROLLING FOR TWD BACKGROUND
FACTORS.
THE TABLE. TABLE 40, WHICH IS PRESENTED IN DB-89,
PRESENTS A COMPARABLE SORT OF PICTURE AFTER CONTROLLING FOR A
MUCH LARGER GROUP OF BACKGROUND FACTORS, 14 BACKGROUND FACTORS
THAT HAVE AN IMPORTANT IMPACT IN THE SYSTEM. SQ THAT ONE CAN
VIEW EACH OF THESE LEVELS ON THIS SCALE HERE FROM ONE TQ EIGHT
AS COMPARABLE TO THE. TO THESE SUB-GROUPS OF CASES IN FIGURE 2,
87?
BALIUS - DIRECT
fo
l | AFTER YOU CONTROL FOR THESE VARINUS BACKGROUND FACTORS, YOU SEE
i
ry
THE CASES BECOMING MORE AND MORE AGGRAVATED AS MEASURED BY THE
WAY THE SYSTEM SEEMS TO BE RESPONDING TO THEM.
H
W
AND YOU CAN SEE THAT THE DEATH SENTENCING RATES IN
THESE LOW LEVEL CATEGORIES FROM ONE TQ SIX ARE VERY SMALL.
THERE ARE VIRTUALLY NO DEATH SENTENCES IN THESE CATEGORIES OF
CASES.
1
EE
G
TH
S
R
L
7
A. YOU RE REFERRING AT THIS POINT TO DB-8%7
v A. CORRECT. THE LOW POINTS ON THE SCALE. ONE THROUGH SEVEN, IN
10 FACT, THE DEATH SENTENCING RATE IN THOSE CATEGORIES NEVER RISES
11 ABOVE, ONLY IN ONE CASE DOES IT RISE ABOVE TWO PERCENT. iz THE DEATH SENTENCING CASES ARE RESERVED FOR THIS TOP
13 |CATEGORY.
14 S50 AS A CONSEQUENCE. WE CAN SEE FROM THIS THAT THERE
13 ARE RELATIVELY FEW CASES IN THE ENTIRE SYSTEM THAT ARE GENUINELY
16 AT RISK WITH RESPECT TO RECEIVING A DEATH SENTENCE. AND THOSE
17 ARE THE CASES THAT ARE CLASSIFIED HERE IN CATEGORY 8.
18 AND ONE STEP OF THE ANALYSIS WAS TO ATTEMPT TO ESTIMATE
Ww 19 THE RACIAL DISPARITIES WITHIN THESE SUB-GROUPS. AND THAT WAS
20 DONE BY JUST SELECTING OUT THE CASES IN EACH OF THESE SUB-CGROUPS
21 WHERE THERE WERE DEATH SENTENCING RATES OF ANY TYPE, AND
rar DETERMINING WHETHER WITHIN THIS CLASS OF CASES THAT WE SAW ANY
23 RACIAL EFFECTS AT ALL WHEN WE DID IT. A MULTIPLE REGRESSION
24 ANALYSIS, AND THOSE ARE THE RESULTS THAT ARE REPORTED IN COLUMNS
ors C AND D OF DB-89.
a
]
230
BALDUS - DIRECT
1 AND THE SHORT OF IT IS YOU CAN SEE THE RACIAL EFFECTS
2 ARE CONCENTRATED IN THIS CATEGORY WHERE THE DEATH SENTENCING
3 'RATE IS ELEVATED: REFLECTS THE SAME THEME. IF YOU HAVE A LOW
4 DEATH SENTENCING RATE. YOURE VERY UNLIKELY TO HAVE MUCH OF A
bo! RACIAL DISPARITY APPARENT, AND THAT PATTERN IS REPEATED HERE.
& THIS IS A CATEGORY IN WHICH YOU FIND 107 OF ALL 128
7 DEATH SENTENCES THAT ARE INCLUDED IN OUR SAMPLE.
8 I. YOU MEAN CATEGORY 8%
4 A. I’M SORRY. I MEANT CATEGORY 8.
10 @. NOW, DID YOU DO ANY ANALYSIS TO FURTHER SUBDIVIDE OR SORT
11 OUT THESE CASES BY A LEVEL OF AGGRAVATION TO CONTINUE THIS KIND
12 OF ANALYSIS?
13 A. YES. THE PRIOR RESEARCH WOULD, SUGGESTED THAT IT WAS
14 |PROBABLY INAPPROPRIATE TO TREAT THE CASES IN CATEGORY 8, SAY, AS
15 ‘A STRICTLY COMPARABLE GROUP OF CASES. THERE PROBABLY WOULD BE A
i& LOT OF VARIATION IN THOSE CASES IN TERMS OF CHARACTERISTICS THAT
37 WERE USED TO SORT OUT THE CASES THAT ARE IN DB-29.
18 S00 WHAT WE DID WAS TO FOCUS IN ON CASES THAT LOCKED
WW 19 LIKE THEY WERE THE BEST CANDIDATES TO RECEIVE A DEATH SENTENCE,
20 GIVEN THE FACTS OF THEIR CASES. AND THAT RESULTED IN THE
21 IDENTIFICATION OF APPROXIMATELY FIVE HUNDRED CASES. 1 MADE THAT
22 JUDGMENT ON THE BASIS OF LOOKING AT THIS, THIS TABLE.
23 R. WHICH TABLE DO YOU MEAN?
24 A. DB-8% AND 1 SAID WELL, THE TOP TWO LEVELS THERE IS PROBABLY
2% WHERE WE-“LL SEE THE BEST SENSE OF WHERE THE LIKELIHOOD OF A
agsl
BAL.DUS - DIRECT
1 DEATH SENTENCE TAKES HOLD, UP TO THE PQINT OF WHERE THE
2 LIKELIHOOD OF A DEATH SENTENCE IS VERY HIGH.
3 AND I FOCUSED IN ON THOSE CASES, AND IF YQU WILL. SORT
OF ENLARGED THE BREAKDOWN WITHIN THOSE CASES. EXPANDED THE
SUBDIVISION OF CASES WITHIN THAT SUB-GROUP.
@. AND LET ME DIRECT YOUR ATTENTION TO DB-90 MARKED FOR
4
5
&
? IDENTIFICATION. CAN YOU IDENTIFY THAT DOCUMENT?
8 A. YES. DB-90 IS THE RESULT OF THAT ENLARGEMENT OF THAT TOF
7 CATEGORY OF APPROXIMATELY FIVE HUNDRED CASES.
0 @. YOU’RE TALKING OBVIOUSLY NOT ABOUT AN ENLARGEMENT OF NUMBER
11 OF THOSE CASE, BUT AN ENLARGEMENT OF A MICROSCOPIC VIEW, IF YOU
12 WOULD, OF HOW THOSE CASES ARE SORTED?
13 A. YES. PUTTING IT MORE PRECISELY. RA FURTHER SUBDIVISION OF
14 THE CASES ALONG THE SAME INDEX THAT WAS USED TO CREATE DB-89.
1% |Q. WHAT DOES DBE~®0 REVEAL?
146 A. DB-90 PRESENTS THE RACE OF VICTIM DISPARITIES CONTROLLING
17 FOR THE RACE OF THE DEFENDANT.
18 AND WHAT WE SEE HERE IS IN COLUMN B, COLUMN A. AN
o 1? IDENTIFICATION WHICH IS PRESENTED IN COLUMN A OF THE LEVELS OF
20 PREDICTED LIKELIHOOD, THAT IS. WE GO FROM LEVEL ONE TO EIGHT,
21 FROM THE LEAST LIKELY PREDICTED DEATH SENTENCE TO THE MOST
22 LIKELY PREDICTED DEATH SENTENCE, GIVEN THE FACTORS THAT WERE
23 BEING USED. THESE ARE THE FACTORS THAT TEND TO EXPLAIN THE
24 RESULTS.
23 AND YOU CAN SEE IN COLUMN B THE ACTUAL DEATH SENTENCING
BALDUS - DIRECT
>
RATE AMONG THESE SUB-CATEGORIES OF CASES. AND THESE
BB CATECORIES WERE DEFINED IN TERMS OF EGUAL NUMBERS OF CASES.
'1I TRIED TO MAKE THE CATEGORIES EQUAL. BUT BECAUSE OF THE. WHERE
THE SCORES BROKE, SOMETIMES THE CATEGORIES WERE NOT THE SAME AND
I WANTED TO KEEP EACH CASE WITH AN IDENTICAL SCORE IN THE SAME
CATEGORY. THAT'S WHY THE CATEGORIES VARY AS THEY DO IN SIZE.
AND I CALCULATED FIRST THE DEATH SENTENCING RATE AMONG
| THOSE CATEGORIES. SO YOU CAN SEE EVEN WITHIN THIS TOP FIVE
HUNDRED GROUP OF CASES THAT STILL THE DEATH SENTENCING RATES DO
T
y
ZY
N
G
0
~
A
B
W
NOT BEGIN TO RISE TO A VERY HIGH LEVEL UNTIL YOU GET UP INTO
[W
Y
Fo
y THE TOP FOUR CATEGORIES. AND EVEN THERE THEY ARE NOT VERY HIGH.
12 IT’S NOT REALLY UNTIL YOU GET UP INTO THE VERY HIGHEST
13 CATEGORY, EIGHT, THAT THE DEATH SENTENCING RATE SHARPLY RISES.
14 ILL MENTION To You, ALTHOUGH IT’S NOT REPORTED HERE. THAT AMONG
15 THAT TOP 58, IF YOU LOOK AT THE TOP 40 IN THAT CATEGORY, MY
16 RECOLLECTION IS THAT EVERY ONE OF THOSE OFFENDERS RECEIVED THE
17 DEATH SENTENCE. THOSE ARE THE MOST AGGRAVATED CASES.
18 NCW, THIS TABLE PRESENTS IN C AND D THE DEATH
WW 19 SENTENCING RATE AMONG BLACK AND WHITE VICTIM CASES. IN THOSE
20 CASES INVOLVING A BLACK DEFENDANT.
21 AND ONE CAN SEE THAT IN THE LOWER CATEGORIES HERE WHERE
<2 THE DEATH SENTENCING RATE IS NOT ELEVATED, THAT THE RACIAL 23 DISPARITIES ARE NOT GREAT BECAUSE THERE ARE NO, THERE’S NO DEATH
24 SENTENCING RATE BASICALLY.
23 BUT ONCE THE DEATH SENTENCING RATE BEGINS TO RISE.
BALDUS - DIRECT
| YOU“LL NOTE THAT IT RISES FIRST IN THE WHITE VICTIM CASES. IT
RISES THERE MORE SHARPLY THAN IT DOES IN THE BLACK VICTIM CASES.
THAT CAN BE SEEN BY COMPARING. IF YOU MOVE DOWN C AND D,
COMPARING THE COMPARABLE FIGURES AT EACH LEVEL ALONG THE WAY.
YOU CAN SEE THAT THE NUMBERS RISE MORE SHARPLY AT EACH STEP
ALONG THE WAY IN COLUMN C THAN THEY DO IN COLUMN D. AND WHEN
YOU MAKE A COMPARISON OF THOSE NUMBERS. AS WE HAVE DONE IN E AND
F, YOU GET SOME MEASURE OF THE RACE OF VICTIM DISPARITY.
AND THE OTHER FEATURE OF THIS IS THAT WE START OUT WITH
SUBSTANTIALLY EQUAL TREATMENT IN THE CASES WHERE THE DEATH
SENTENCING RATE IS LOW, THE DEATH SENTENCING RATE RISES, AND THE
DISPARITIES RISE.
BUT ONCE THE CASES BECOME QUITE AGGRAVATED. THATS UP
IN THIS CATEGORY 8, THE DEATH SENTENCE. THE DEATH SENTENCING
RATE GOES UP QUITE HIGH, BUT THE DISPARITIES BEGIN TO DECLINE.
AND WE-LL NOTE HERE IN THIS TOP CATEGORY, IT DROPS DOWN TO .16.
‘BUT AS 1 MENTIONED EARLIER, IF YOU FURTHER SUBDIVIDED THAT YOU
WOULD FIND THAT AMONG THE TOP FORTY OF THESE CASES. THE DEATH
|
SENTENCING RATE IS THE SAME FOR THE BLACK AND THE WHITE VICTIM
CASES, BECAUSE IN THAT CATEGORY, EVERYBODY IS GETTING A DEATH
SENTENCE.
| WHAT THESE NUMBERS SUPPORT IS THE HYPOTHESIS THAT'S
BEEN DEVELOPED IN EARLIER RESEARCH, MOST NOTABLY IN THE WORK OF
HARRY CALVIN, AND HANS ZEIZEL WHO FUBLISHED A VERY EXTENSIVE
ANALYSIS OF JURY DECISION MAKING IN A BOOK CALLED "THE AMERICAN
|
834
BALDUS ~ DIRECT
i JURY." IT WAS PUBLISHER IN 1946.
2 | AND IN THAT BOOK, THEY DEVELOPED A HYPOTHESIS, A TESTED
HYPOTHESIS, WHAT THEY CALL THE LIBERATION HYPOTHESIS AND IN
SHORT WHAT IT WAS, THAT THE EXERCISE OF DISCRETION IS
CONCENTRATED IN THE AREA WHERE THERE-“S REAL ROOM FOR CHOICE. IF
2
4
3
iB & THE IMPOSITION OF A PENALTY OR A CONVICTION IS NOT THINKABLE
7 BECAUSE THE. EVIDENCE IS JUST NOT THERE. THOSE ARE THE CASES THEY
8 ‘WERE DEALING WITH, GUILT CONVICTIONS AT TRIAL, OR THE EVIDENCE
? IS OVERWHELMING, THAT NO REASONABLE PERSON COULD DISAGREE THAT
10 THIS PERSON COULD BE CONVICTED, THEY SAW NO RACIAL EFFECTS, NO
11 EFFECTS OF ANY ARBITRARY FACTORS IN THE DECISION PROCESS.
12 i BUT WHEN YOU LOOK AT THE CASES IN THE MID, WHAT THEY
13 CALL THE MID-RANGE. WHERE THE FACTS DO NOT CALL CLEARLY FOR ONE
14 CHOICE OR ANOTHER, THAT THATS WHERE YOU SEE THERES ROOM FOR
15 THE EXERCISE OF DISCRETION. THEY CHARACTERIZE IT, IF YOU WILL.
146 THE FACTS LIBERATE THE DECISION MAKER TO HAVE A BROADER FREEDOM
17 FOR THE EXERCISE OF DISCRETION, AND IT IS IN THE CONTEXT OF
13 THOSE DECISIONS THAT YOU SEE THE EFFECTS OF ARBITRARY OR
Ww 19 POSSIBLY IMPERMISSIBLE FACTORS WORKING.
20 AND THAT’S THE PATTERN WE SEE HERE IN THIS TABLE, WHERE
21 THE RACE. THE DISPARITIES RISE AS WE MOVE INTO THIS GREY AREA,
<2 WHERE THE FACTS ARE NOT CLEAR, AND THE EVIDENCE THAT THE FACTS
23 ARE NOT CLEAR ON CULPABILITY IS THAT AMONG THESE CLASSES OF
24 CASES THE DEATH SENTENCING RATE IS NOT HIGH.
23 BUT ONCE THE CASES BECOME QUITE AGGRAVATED. THEY REACH
N
S
E
S
O
R
E
N
SR
R
t
RE
E
10
11
883
BALDUS -~ DIRECT
i
|
A POINT WHERE NO REASONABLE PERSON COULD DISAGREE THAT IF WE'RE
GOING TO HAVE A DEATH SENTENCE, THESE ARE THE CASES THAT DESERVE
IT. THEN THE DISPARITIES DISAPPEAR.
MR. BOGER: YOUR HONOR, AT THIS TIME, I MOVE THE
ADMISSION OF DB-8% AND DB-?0 INTO EVIDENCE.
MS. WESTMORELAND: SAME OBJECTION.
THE COURT: ALL RIGHT, THEY LL BE ADMITTED.
MR. BOGER: THANK YOU. YOUR HONOR.
BY MR. BOGER:
Q. PROFESSOR BALDUS. HAVE YOU DONE THE SAME SORT OF ANALYSIS
WITH RESPECT TO THE RACE OF DEFENDANT DISPARITIES. AMONG THESE
HIGHLY AGGRAVATED CASES?
A. YES. THE, THE, A COMPARABLE ANALYSIS, YOUR HONOR, FOR THE
RACE OF DEFENDANT DISPARITIES IN THE CONTEXT QF WHITE VICTIM
CASES, IS SHOWN IN DB-91, AND IT SHOWS THE SAME SORT OF PATTERN.
BASICALLY, WHAT WE“VE DONE HERE IS RELOCATE THE COLUMNS
'C, D, G AND H THAT WERE PRESENTED IN DB-90. WE JUST REORDERED
THEM SO WE WERE ABLE TO CONTROL FOR THE RACE OF VICTIM AND NOW
MEASURE DISPARITIES IN TERMS OF THE RACE OF THE DEFENDANT. | AND COLUMNS C AND D SHOW THOSE DISPARITIES, AND AS ONE
GOES DOWN COLUMNS C AND D. YOU CAN SEE THAT THE, AS THE CASES
BECOME MORE AGGRAVATED, THE DEATH SENTENCING RATES RISE MORE
SHARPLY IN THE CASES INVOLVING BLACK OFFENDERS THAN THEY DO IN
THE CASES INVOLVING THE WHITE OFFENDERS. AND THAT IS WHAT
PRODUCES THE DISPARITIES THAT ARE APPARENT IN COLUMNS E AND F.
FE
| BALDUS ~- DIRECT
AND AGAIN WE SEE EVIDENCE OF THE LIBERATION HYPOTHESIS
|
|
|
|
|
|
{ i {
1 HERE. THE DISPARITIES ARE WEAK, OR NON-EXISTENT VIRTUALLY IN
MN
3 |THE LESS AGGRAVATED CASES. IN THE CENTRAL RANGE IS WHERE WE SEE
4 THEM. THE MID-RANGE OF CASES IS WHERE THERE IS MORE AMBIGUITY AS
(iL
T0 THE APPROPRIATENESS OF A DEATH SENTENCE. AND THEN THEY
5 A DISAPPEAR IN THE HIGH LEVEL CATEGORY. UP IN THIS CATEGORY WHERE
7 88 PERCENT OF THE PEOPLE RECEIVING THE DEATH SENTENCE. THERES
§ [VIRTUALLY NO DISPARITY. THOSE ARE CASES THAT APPARENTLY ARE
9 |SUCH THAT CONSIDERATIONS OF RACIAL FACTORS ARE JUST OVERWHELMED
10 BY THE FACTS IN THE CASES. AND WE SEE NO EFFECTS IN THOSE
11 CONTEXTS. THAT’S WHAT THE DATA SUGGESTS TO ME.
12 MR. BOGER: YOUR HONOR, I MOVE THE ADMISSION OF DB-91
13 INTO EVIDENCE,
14 THE COURT: IF THERE'S NO ADDITIONAL OBJECTION, IT WILL
15 BE ADMITTED.
16 MS. WESTMORELAND: THE ONES I STATED PREVIOUSLY, YOUR
17 HONOR.
12 THE COURT: IS 82 A DIFFERENT TACT THAN 227?
w 19 MR. BOGER: PARDON?
20 THE COURT: IS 92 A DIFFERENT TACT?
21 MR. BOGER: YES, YOUR HONOR, 92 IS A DIFFERENT TACT.
22 THE COURT: ALL RIGHT, LET’S TAKE A TEN-MINUTE BREAK.
23 ick yn |
24 (RECESS TAKEN.)
23 THE COURT: ALL RIGHT. MR. BOGER, GO AHEAD.
287
BALDUS - DIRECT
Pr
y MR. BOGER: THANK YOU, YOUR HONOR.
DAVID C. BALDUS,
BEING PREVIOUSLY DULY SWORN, RESUMED THE WITNESS STAND AND
TESTIFIED FURTHER AS FOLLOWS: :
DIRECT EXAMINATION (CONT D)
BY MR. BOGER:
®t. PROFESSOR BALDUS, DID YOU AT ANY POINT UNDERTAKE ANY
EY
H
R
B
E
1
S
E
p
a
Ln
©
Da
d
ANALYSIS TO DETERMINE WHY THESE CASES WERE BEING TREATED
10 DIFFERENTLY?
11 THE COURT: BEING WHAT?
12 MR. BOGER: BEING TREATED DIFFERENTLY.
13 BY MR. BOGER?
14 QR. AT THESE DIFFERENT POINTS IN THE SYSTEM, OR HOW THAT WAS
13 HAPPENING?
1& A. WELL, THE ANALYSIS WE JUST LOOKED AT SUFPFORTS THE BROAD
17 CONCLUSION THAT LEVELS OF AGGRAVATION IN CASES THAT ARE
13 SIMILARLY SITUATED ARE EVALUATED DIFFERENTLY IN THE SYSTEM.
ww 1% @. LET ME DIRECT YOUR ATTENTION TO DB-92 MARKED FOR
20 IDENTIFICATION.
=1 CAN YOU IDENTIFY THAT DOCUMENT?
22 Ae YES, DB-92 PRESENTS THE RESULTS OF AN ANALYSIS WHICH WAS
23 DESIGNED TO FOCUS MORE SHARPLY ON WHY WE WOULD OBSERVE. OR TO
24 DEVELOP MORE EVIDENCE ON THE QUESTION OF WHETHER DR NOT THE
2% [FACTORS THAT TEND TO AGGRAVATE CASES ARE INDEED EVALUATED
288
BALDUS - DIRECT
1 DIFFERENTLY IN WHITE VICTIM AND BLACK VICTIM CASES.
[ AND THE WAY THIS WAS DONE WAS TO IDENTIFY THE CASES
THAT APPEAR TO BE IMPORTANT IN THE SYSTEM. AND THAT WAS THE. WE
W
w
W
STARTED WITH A LIST OF 39 VARIABLES THAT WE PRESENTED IN A TABLE
3 YESTERDAY.
Ri & 2. EXCUSE ME, PROFESSOR BALDUS. YOU SAID TO IDENTIFY THOSE
7 CASES THAT WE IDENTIFIED AS IMPORTANT IN THE SYSTEM. DO YOU
a {MEAN -—
? A. YES, EXCUSE ME, I MEAN TQ IDENTIFY THE VARIABLES THAT APPEAR
0 TO BE IMPORTANT IN THE SYSTEM.
11 AND WE STARTED OUT WITH THE HYPOTHESIS THAT IF THE
12 SYSTEM WERE INDEED EVALUATING IMPORTANT AGGRAVATING AND
13 MITIGATING FACTORS THE SAME, THAT WOULD TEND TO MEAN THAT THE
14 CASES WERE BEING, CASES THAT WERE SIMILAR WOULD BE TREATED THE
15 SAME.
16 IF INDEED COMPARABLE CASES. IN TERMS OF AGGRAVATING AND
17 MITIGATING FACTORS. WERE BEING TREATED THE SAME, THEN WITHIN A
13 CATEGORY OF WHITE VICTIM CASES YOU WOULD EXPECT THOSE
Ww 19 AGGRAVATING FACTORS TO HAVE AN EFFECT ON THE DISPOSITION OF
20 BLACK VICTIM CASES TO BE COMPARABLE TO THE EFFECT THEY WOULD
at HAVE ON WHITE VICTIM CASES. 350 WE SEPARATED THE DATA SET INTO
2% TWO POPULATIONS. ONE, THE POPULATION OF WHITE VICTIM CASES AND
23 THE OTHER THE POPULATION OF BLACK VICTIM CASES. AND CONDUCTED 4
24 SEPARATE MULTIPLE REGRESSION ANALYSIS WITHIN EACH POPULATION.
23 AND IDENTIFIED THE AGGRAVATING FACTORS THAT APPEARED TO BE
BALDUS - DIRECT
1 IMPORTANT IN EACH OF THOSE ANALYSES.
2 WE THEN DETERMINED THE VARIABLES THAT WERE AGGRAVATING
3 THAT APFEARED TO BE MORE IMPORTANT IN WHITE VICTIM CASES THAN
4 WAS THE CASE IN THE BLACK VICTIM CASES AND AS A RESULT OF DOING
Rh
THAT SORT, WE FOUND THAT MANY MORE OF THE AGGRAVATING FACTORS
R 6 THAT APPEARED TO BE IMPORTANT IN THE SYSTEM, PLAYED A MORE
7 IMPORTANT ROLE IN THE WHITE VICTIM CASES THAN THEY DID. IN THE
0)
BLACK VICTIM CASES, AND DB-92 PRESENTS THE RESULTS OF THIS
2 ANALYSIS.
10 AND IT SHOWS IN COLUMN B THE MAGNITUDE AND STATISTICAL
BISNIEICANCE OF THE CO-EFFICIENTS THAT WERE ESTIMATED, USING A
i2 LEAST SQUARES REGRESSION ANALYSIS FOR THE AGORAVATING FACTORS.
13 THE COLUMN C SHOWS A SIMILAR SORT OF ANALYSIS WITH
i4 RESPECT TO THOSE SAME AGGRAVATING FACTORS AMONG THE BLACK VICTIM
15 | CASES.
16 | AND WHEN ONE READS DOWN COLUMN B, YOU CAN SEE THAT A
17 GOODLY NUMBER OF THOSE CASES ARE, CORRECTION, A GOODLY NUMBER OF
18 THOSE VARIABLES ARE LARGE AND STATISTICALLY SIGNIFICANT, WHEREAS
& Poy AMONG THE BLACK VICTIM CASES, THE CO-EFFICIENTS ARE SMALLER, AND
20 ARE LESS LIKELY TO ACHIEVE STATISTICAL SIGNIFICANCE.
| TO MAKE THE COMPARISON SOMEWHAT MORE PRECISE. WE
|
|
|
}
<3 REGRESSION CO-EFFICIENTS, AND ALSO CALCULATED THE
CALCULATED THE DIFFERENCE IN THOSE. THE MAGNITUDE OF THOSE
24 STATISTICAL SIGNIFICANCE OF THE DIFFERENCE IN THE MAGNITUDE OF
23 THOSE RATES, AND THOSE RESULTS ARE PRESENTED IN COLUMNE D AND
290
BALDUS - DIRECT
E. D SHOWS THE ARITHMETIC DIFFERENCE. D SHOWS, CORRECTION. E
SHOWS THE RATIO.
|
|
AND AS YOU CAN SEE, THESE AGGRAVATING FACTORS THAT ARE
LISTED AT THE TOP OF THIS TABLE, HAVE A MUCH MORE SUBSTANTIAL
IMPACT IN THE WHITE VICTIM CASES THAN THEY DO IN THE BLACK
VICTIM CASES.
@. WHAT DOES THAT MEAN IN TERMS OF THE WAY THE SYSTEM MIGHT
OPERATE?
A. THAT MEANS IF YOU LOOK AT A SET OF CASES THAT ARE COMPARABLE
IN TERMS OF THESE STATUTORY AGGRAVATING FACTORS AND OTHER |
NON-STATUTORY AGGRAVATING FACTORS, THAT THE SYSTEM RESPONDS MORE
SHARPLY TO THE PRESENCE OF THESE FACTORS IN WHITE VICTIM CASES
THAN IT DOES IN BLACK VICTIM CASES.
THATS WHAT THESE DISPARITIES IN THESE REGRESSION
CO-EFFICIENTS REFLECT. IT TELLS YOU IN SOMEWHAT MORE DETAIL THE
SAME STORY THAT DB-90 AND DB-%?1 TOLD YOU. THERE, THE INDEX THAT
WAS USED AGGREGATED THE COMBINED EFFECTS OF ALL OF THESE
VARIABLES TO SPECIFIED GROUPS OF SIMILARLY SITUATED CASES, AND
WE SAW DIFFERENT DEATH SENTENCING RESULTS AMONG THOSE DIFFERENT
GROUPS OF COMPARABLE CASES.
HERE THE FOCUS IS ON THE SPECIFIC AGGRAVATING FACTOR.
AND THIS ANALYSIS PROVIDES EVIDENCE THAT IN WHITE VICTIM CASES,
THESE SPECIFIC AGGRAVATING FACTORS COUNT MORE AGAINST A
DEFENDANT THAN THEY DO IN WHITE VICTIM CASES. AND IT JUST
PROVIDES FURTHER DETAIL ON THE SOURCES OF THE DISPARITIES THAT
[3
[8
BALDUS - DIRECT
WE SEE WHEN WE EXAMINE THE OVERALL RACE OF VICTIM EFFECTS IN THE
SYSTEM.
@. BUT I THOUGHT YOU TESTIFIED EARLIER THAT IN GENERAL, WHITE
VICTIM CASES ARE MORE AGGRAVATED THAN BLACK VICTIM CASES.
‘A. THAT”S TRUE. WHITE VICTIM CASES ARE SUBSTANTIALLY MORE
AGGRAVATED THAN BLACK VICTIM CASES.
WHAT THIS ANALYSIS SHOWS ONE IS THAT EVEN WHEN THE SAME
FACTORS ARE PRESENT IN THE CASES, HOWEVER, WHEN THE CASES ARE
COMPARABLE, THAT THE SYSTEM RESPONDS DIFFERENTLY TO THE CASES
THAT HAVE AGGRAVATING FACTORS WHEN THE VICTIM IS WHITE AND WHEN
IT IS BLACK.
THE SOURCE OF THE DISPARITIES THAT WERE SEEN IN THE
SYSTEM IS THE TREATMENT OF CASES THAT ARE COMPARABLE WITH
RESPECT TO THE KEY AGGRAVATING AND MITIGATING FACTORS IN THE
SYSTEM.
AND IT IS TRUE, TO BE SURE, THAT ON THE WHOLE. THE
WHITE VICTIM CASES ARE SUBSTANTIALLY MORE AGGRAVATED, BUT THAT
IS NOT WHAT PRODUCES THE RACE OF VICTIM EFFECTS THAT WE SEE.
IT’S THE WAY THE SYSTEM TREATS THOSE GROUPS OF CASES THAT ARE
COMPARABLE WITH RESPECT TO AGGRAVATING AND MITIGATING FACTORS
AND INDEED THERE ARE SUBSTANTIAL NUMBERS OF CASES IN THE SYSTEM
(WHICH ARE COMPARABLE IN TERMS OF THESE FACTORS. AND THATS THE
SOURCE OF THE RACE OF VICTIM EFFECTS THAT WE SEE PRODUCED BY OUR
MEASURES.
MR. BOGER: YOUR HONOR, AT THIS TIME. I MOVE DB-%2 INTO i
BALDUS - DIRECT
1 EVIDENCE.
THE COURT: LET ME ASK YOU A QUESTION, PROFESSOR
BALDUS.
IF I UNDERSTAND WHAT YOU JUST TESTIFIED TO. YOU HAVE
3
4
= SELECTED THE VARIABLES IN COLUMN A, BASED ON AN
& ANALYSIS OF THE WHITE VICTIM CASES. AND YOU HAVE FOUND THAT
7 THESE ARE IN YOUR VIEW, STRONG MOTOR POWER FOR SELECTION OF A
3 DEATH PENALTY.
? THE WITNESS: NO, THEY WERE BASED ON ANALYSIS GF ALL
10 ‘THE CASES. YOUR HONOR.
11 THE COURT: THAT'S WHAT I MISUNDERSTOOD.
12 THE WITNESS: THESE ARE THE VARIABLES THAT WERE
13 REFLECTED IN THE ANALYSIS PRESENTED YESTERDAY, YOUR HONOR, IN
14 TABLES 31A, AND 31B. I INDICATED YESTERDAY THAT WE STARTED WITH
135 ‘A POOL OF 3% VARIABLES AND BOILED THEM DOWN. THIS IS THE SAME
16 GROUP OF THOSE VARIABLES.
17 MR. BOGER: THAT’S DB~81 AND DB-22, IS THAT CORRECT?
18 THE WITNESS: YES, THAT'S RIGHT.
Ww 1? ANOTHER THING THAT WAS INTERESTING TD ME ABOUT THIS
20 ANALYSIS, YOUR HONOR. WAS THAT WE LOOKED AT THE WAY THE SYSTEM
21 EVALUATES MITIGATING FACTORS. AND THAT PART OF THE ANALYSIS IS
22 PRESENTED ON THE LAST PAGE OF DB-%2,
23 ; AND IF YOU-LL LOOK ON THAT PAGE, YOUR HONOR, IT”3 PART
24 24, LAST PAGE OF THIS EXHIBIT, YOU SEE HERE, WE HAVE THE. WE
23 HAVE LISTED HERE THE MITIGATING FACTORS THAT, WHEN EXAMINED IN
BALDUS -~ DIRECT
1 | THE CONTEXT OF ALL THE CASES, HAVE THE MOST STATISTICAL EFFECT
2 | IN PREDICTING DEATH SENTENCES.
3 AND WE DID THIS SAME ANALYSIS WITH THESE VARIABLES. IN
FACT THEY WERE INCLUDED IN THE ANALYSIS WITH THE AGGRAVATING
AND AS YOU SEE, AS YOU GO DOWN COLUMNS B AND C OF PART
&
3
= FACTORS THAT WE JUST LOOKED AT.
6
7 2A OF DB-92, YOU 3EE THAT THE MITIGATING FACTORS APPEAR TO HAVE
A MORE SIGNIFICANT EFFECT IN THE WHITE VICTIM CASES THAN IN THE 0
? BLACK VICTIM CASES.
10 THE EVIDENCE HERE IS PLAIN, THAT THRQUCHOUT THE SYSTEM.
11 THE MITIGATING FACTORS HAVE MUCH LESS INFLUENCE IN DETERMINING
12 RESULTS THAN DO THE AGGRAVATING FACTORS. BUT AMONG THOSE THAT
13 DO HAVE AN EFFECT, WHAT IS STRIKING HERE, I THINK, IS THAT THE
14 EFFECT IS STRONGER IN THE WHITE VICTIM CASES. AND I PUZZLED
15 {OVER THAT. AND IT, IT APPEARS TO ME THAT WHAT IT SUGGESTS IS
16 THAT WHEN THE SYSTEM IS DEALING WITH BLACK VICTIM CASES. THAT
17 THEY SIMPLY DO NOT RECEIVE THE SAME LEVEL OF SCRUTINY THAT THE
18 WHITE VICTIM CASES RECEIVE. THAT THE AGGRAVATING FACTORS DON’T
w Bc COUNT AS MUCH, AND SIMILARLY, THE MITIGATING FACTORS DON'T
20 APPEAR TO COUNT AS MUCH WHEN THOSE CASES ARE PROCESSED THROUGH
THE SYSTEM.
| MR. BOGER: YOUR HONOR, ILL RENEW MY MOTION TO ADMIT
23 |pB-92.
24 THE COURT: ALL RIGHT. UNLESS THERE‘S ANY ADDITIONAL
ite OBJECTION, IT WILL BE ADMITTED.
ad
u
o
NM
e
G
h
W
o
N
10
11
2324
BALDUS - DIRECT
MS. WESTMORELAND: NO ADDITIONAL OBJECTION.
| MR. BOGER: THANK YOU, YOUR HONOR.
BY MR. BOGER:
@. PROFESSOR BALDUS WE TALKED EARLIER ——
| THE COURT: WAIT JUST A SECOND. GO AHEAD.
MR. BOGER: THANK YOU. YOUR HONOR,
BY MR. BOGER:
‘RQ. PROFESSOR BALDUS. WE TALKED EARLIER, AND I BELIEVE THE JUDGE
SPECIFICALLY ASKED YOU ABOUT THE WAY IN WHICH RACIAL DISPARITIES
IN TREATMENT MIGHT BE EXHIBITED AT DIFFERENT STAGES OF THE
CHARGING AND SENTENCING SYSTEM.
DID YOU DO ANY ANALYSES THAT EXAMINED RACIAL
DIFFERENCES OR DISPARITIES AT DIFFERENT STAGES OF THE SYSTEM?
A. YES.
'R. LET ME ASK YOU TO TURN TO DB-93 MARKED FOR IDENTIFICATION
AND ASK You TO IDENTIFY THAT DOCUMENT. IF YOU CAN?
A. YES. DB-93 IS TABLE 43 FROM OUR REPORT, AND IT PRESENTS THE
PROPORTIONS OF WHITE VICTIM CASES ——
(0. EXCUSE ME, PROFESSOR BALDUS. DID YOU SAY DB-43 DR —-
A. I’M SORRY, IF I DID. NO, ITS TABLE 43B, FROM THE TABLE
WHICH IS DB-93.
IT PRESENTS THE PROPORTION OF WHITE VICTIM CASES THAT
ARE FOUND AT EACH LEVEL OR AT EACH STAGE IN THE PROCESS AS THE
CASES PROCEED FROM POINT OF INITIAL INDICTMENT TO DEATH ROW.
AND TABLE, PARDON ME, COLUMN B SHOWS THOSE PROPORTIONS.
hd
5
3
BALDUS - DIRECT
IT INDICATES THAT 39 PERCENT, OR .3% OF THE OFFENDERS INDICTED
FOR MURDER OR VOLUNTARY MANSLAUGHTER HAD WHITE VICTIMSZ.
OF THOSE INDICTED FOR MURDER, .40 HAD WHITE VICTIMS. SO
ON AND S0 FORTH, UNTIL YOU REACH THE POPULATION OF OFFENDERS
SENTENCED TO DEATH. AND ONE CAN SEE THAT THE PROPORTION OF WHITE
VICTIMS HAS RISEN TO .84.
COLUMN C PRESENTS A SIMILAR ANALYSIS WITH RESPECT TO
THE PROPCRTION OF CASES INVOLVING BLACK DEFENDANTS AND WHITE
VICTIMS, AT SUCCESSIVE STAGES, AND THERE YOU CAN SEE BY
COMPARING ROWS ONE AND EIGHT THAT THE PROPORTION OF CASES
INVOLVING BLACK DEFENDANTS AND WHITE VICTIMS, ELEVATES FROM ,09
AT THE POINT OF INDICTMENT TO .39 TO THE, IN THE POPULATION OF
OFFENDERS SENTENCED TO DEATH.
THE PURPOSE OF THIS ANALYSIS IS TO GIVE ONE AN OVERVIEW
oF THE RATE AT WHICH WHITE AND BLACK VICTIMS RESPECTIVELY DROP
QUT OF THE SYSTEM AS THE CASES ARE WINNOWED AND MOVED TO THE
DEATH SENTENCING HEARING.
THERE“S NO BACKGROUND CONTROLS IN THIS ANALYSIS,
WHATEVER.
THE ANALYSIS DOES, HOWEVER, -—-—
THE COURT: LET ME ASK YOU A QUESTION.
COLUMN C SAYS PROPORTION OF CASES WITH BLACK DEFENDANTS
|
i
AND WHITE VICTIMS.
THE WITNESS: RIGHT.
THE COURT: SO THIS DOESN/T SHOW THE DROPOUT OF BLACK
BALDUS - DIRECT
1 VICTIMS, DOES IT?
2 THE WITNESS: WELL. COLUMN B DOES, YOUR HONOR. WHAT
3 £4 DOING HERE IN THIS TABLE IS FOCUSING ON TWO GROUPS. THE
4 COLUMN B SHOWS AT THAT .39, CORRECTION. .3% OF THE CASES AT THE
3 POINT OF INDICTMENT. INVOLVE WHITE VICTIMS AND QTHERS WERE BLACK
b VICTIMS. AND THERES A, THE COMPLEMENT TO EACH ONE OF THESE IS
7 THE DIFFERENCE BETWEEM THIS PROFORTION AND ONE.
HOWEVER, THAT DOES NOT HOLD TRUE OVER HERE WITH RESPECT
v
o
©
TO COLUMN C. THERE WE FOCUSED ON ONE OF THE FOUR
10 | DEFENDANT-VICTIM RACIAL COMBINATIONS, THAT IS, BLACK DEFENDANT
11 |AND WHITE VICTIM. BUT YOU CAN SEE THAT IN CONTRAST TO THE OTHER
12 RACIAL COMBINATIONS OF DEFENDANTS AND VICTIMS THAT BLACK
13 DEFENDANTS WITH WHITE VICTIMS TEND TO REMAIN IN THE SYSTEM AT A
14 HIGHER RATE THAN DO THE OTHER COMBINATIONS.
15 IN A PERFECTLY EVEN HANDED SYSTEM. WHERE ALL CASES WERE
16 EQUALLY DEATH ELIGIBLE, ONE WOULD EXPECT THAT THE PROPORTION OF
17 |CASES AT EACH STAGE WITH DIFFERENT RACIAL CHARACTERISTICS WOULD
13 |BE THE SAME.
W 19 |BY MR. BOGER!
20 |@. HAVE YOU DONE ANY MORE REFINED ANALYSIS OF THE
21 |VICTIM-DEFENDANT RACIAL BREAKDOWNS AT THE VARIOUS STAGES OF THIS
22 [CHARGING AND SENTENCING SYSTEM?
23 1A. YES.
24 '@. LET ME DIRECT YOUR ATTENTION TO WHAT’S BEEN MARKED AS DE~94
25 AND ASK YOU TO IDENTIFY THAT DOCUMENT?
BALDUS - DIRECT
1 ‘A. DE-24, WHICH IS TABLE 44 IN OUR REPORT, PRESENTS A
2 DISTRIBUTION OF THE RATES OF SELECTION OF OFFENDERS FOR CERTAIN
DISPOSITIONS AT SIX SUCCESSIVE STAGES IN THE PROCESS.
IN ADDITION, WELL, ROW 7 SHOWS THE COMBIMED EFFECTS OF
ALL THE DECISIONS IN THE SYSTEM. AND THE PURPOSE OF THIS TABLE
3
4
5
& & 1S TO PROVIDE MORE DETAIL ON THE WAY IN WHICH CASES ARE DISPOSED
v J OF AT DIFFERENT STAGES IN THE FROCESS.
a "AND COLUMN B INDICATES THE RATES OF DISPOSITION AT EACH
2 OF THESE VARIOUS STAGES. AND I THINK, I SAY BY WAY OF
10 BACKGROUND AGAIN THAT THESE DATA DO NOT PRESENT ANY MEASURES OF
i DISPARITY AND THERE ARE NO ADJUSTMENTS. IT-S SIMPLY TQ PROVIDE
12 AN OVERVIEW AND TO SUGGEST WHERE YOU MIGHT FIND THE LIKELY
13 SOURCE OF THE RACIAL EFFECT THAT WE SEE AND HAVE SEEN WHEN WE
14 EXAMINE THE COMBINED EFFECTS OF ALL THE DECISIONS IN THE SYSTEM.
15 AND THE FIRST SUGGESTION IS PROVIDED IN ROW 2, WHERE WE
16 LOOK AT THE GUILTY PLEA TO VOLUNTARY MANSLAUGHTER AMONG CASES
17 THAT HAVE BEEN INDICTED FOR MURDER. AND IF YOU SEE THERE. IF YOU
18 LOOK AT THAT ROW OVER IN COLUMNS C. D. E AND F, PARTICULARLY
. i? FOCUSING ON COLUMN F, WE SEE THAT THE, AMONG THE BLACK
20 DEFENDANT-BLACK VICTIM CASES. THAT OVER FIFTY PERCENT OF THOSE
21 CASES MOVE OUT OF THE SYSTEM AT THIS STAGE IN THE FORM OF A
22 GUILTY PLEA TO VOLUNTARY MANSLAUGHTER.
23 | WE SEE ~~
24 @. YOURE REFERRING THERE TO COLUMN E?
25 Ae npn, YES.
399
BALIUS — DIRECT
1 WE SEE THAT IN CONTRAST —--
2 THE COURT: THAT S NOT WHAT COLUMN E SAYS. IS THAT
2 |WHAT’S BOTHERING YOU, MR. BOGER?
3 MR. BOGER: I THOUGHT I HEARD HIM TALK ABOUT BLACK AND
5 BLACK COMBINATIONS IN COLUMN F, AND —-—
& THE COURT: AND MORE THAN FIFTY PERCENT AND -- YOU“RE
7 TALKING ABOUT "E"?
8 THE WITNESS: YES.
? THE COURT: OKAY.
10 THE WITNESS: IF I SAID "B", I APOLOGIZE. YES, “E",
11 THE COLUMN I WAS REFERRING TO. AND WE SEE THAT HALF OF THOSE
12 CASES DROP OUT OF THE SYSTEM AT THIS STAGE, WITH A PLEA TO
13 VOLUNTARY MANSLAUGHTER.
14 THE NEXT POINT, I THINK, THATS WORTH NOTING HERE IS AS
15 BETWEEN BLACK AND WHITE OFFENDERS WHO HAVE WHITE VICTIMS, YOU
16 CAN SEE HERE THAT AMONG THOSE WHO PLEAD OUT OF THE SYSTEM. THIS
17 IS A COMPARISON OF COLUMNS C AND D, IN ROWS 2 AND 3, YOU CAN SEE
18 |THAT THE WHITE OFFENDERS. THE WHITE VICTIMS ARE MUCH MORE LIKELY
a 19 TO GO OUT ON A PLEA TO VOLUNTARY MANSLAUGHTER THAN ARE THE BLACK
20 OFFENDERS WITH A WHITE VICTIM: AND THAT AMONG THE BLACK
21 OFFENDERS WITH A WHITE VICTIM PLEADING OUT, THEY TEND TO BE MUCH
22 MORE LIKELY TO GO OUT ON A PLEA TO MURDER THAN DO THE WHITE <3 OFFENDERS.
24 THE NEXT POINT OF INTEREST THAT SUGGESTS WHAT MAY BE
23 THE SOURCE OF DISPARITIES IS PROVIDED IN ROW 5S, WHICH LOOKS AT
BALDUS - DIRECT
1 THE RATES AT WHICH CASES THAT RESULTED IN A MURDER CONVICTION
2 TRIAL AND ARE ADVANCED TO A PENALTY TRIAL.
3 AND HERE, WE SEE A SHARP DIFFERENCE IN THE RATES AT
4 WHICH BLACK OFFENDERS WHO HAVE BEEN FOUND GUILTY OF MURDERING A
5 WHITE GO TO A PENALTY TRIAL AS CONTRASTED WITH THE RATE IN CASES
& WITH A WHITE DEFENDANT AND A WHITE VICTIM. :
7 ALSO WE CAN SEE THAT THE RATE FOR WHITE VICTIM CASES
3 GENERALLY IS SUBSTANTIALLY HIGHER THAN IT IS FOR BLACK VICTIM
2 CASES.
10 FINALLY, WHEN WE REACH LEVEL 4. WHERE WE LOOK AT THE
12 IMPOSITION OF DEATH SENTENCE AT PENALTY TRIAL, HERE WE SEE A
12 | CHANGE IN THE PRECEDING PATTERN. THAT IS, THAT HERE, WHILE THE
13 OFFENDERS WHO HAVE KILLED A WHITE APPEAR TO BE TREATED THE SAME
14 AS BLACK OFFENDERS, INDEED THE RATE OF DEATH SENTENCE IS A
13 FRACTION LOWER, FOUR PERCENTAGE POINTS LOWER.
16 HOWEVER, WHEN WE COMPARE THE WHITE VICTIM CASES WITH
17 THE BLACK VICTIM CASES, WE SEE A SLIGHT DISPARITY IN THESE
ie UNADJUSTED FIGURES. THAT IS. A COMPARISON OF THE .43 IN COLUMN
iw 19 E FOR THE BLACK DEFENDANT-BLACK VICTIM CASES AS CONTRASTED WITH
20 THE SOMEWHAT HIGHER LEVELS AMONG THE WHITE VICTIM CASES. AGAIN
Z1 I MENTION THESE ARE ALL WITHOUT ADJUSTMENTS FOR BACKGROUND
22 FACTORS.
£3 | THE COURT: WHAT THAT SAYS, IF I UNDERSTAND IT. AND LET
24 ME JUST STATE IT, IS THAT A BLACK DEFENDANT WITH A WHITE VICTIM
Pit) HAS GOT A FOUR PERCENT HIGHER CHANCE OF CETTING THE DEATH
S00
BALDUS — DIRECT
PENALTY IF ADVANCED THAN A WHITE.
AND THEN UNDER E AND F, A BLACK DEFENDANT WITH A BLACK
VICTIM HAS A SEVEN PERCENT LESS CHANCE OF GETTING THE DEATH
PENALTY THAN A WHITE-BLACK, IS THAT RIGHT?
THE WITNESS: I DID NOT -- I LOST THE TRAIN OF WHERE
YOUR HONOR WAS HEADING THERE AT THAT POINT. YOUR HONOR.
THE COURT: I“M LOOKING AT LINE &, I“M NOT GOING —-
THE WITNESS: YES. OKAY.
THE COURT: ~~ ANYWHERE, IM JUST TRYING TO UNDERSTAND
WHATS HERE.
WE“VE GOT A SIXTY PERCENT PROBABILITY RATE ——
THE WITNESS: THATS RIGHT.
THE COURT: ~—- FOR BLACK~-WHITE, AND A 3&6 PERCENT FOR WHITE-WHITE.
| THE WITNESS: THAT’S RIGHT. YOUR HONOR.
THE COURT: THE WHITE HAS A LITTLE BIT OF AN ADVANTAGE
THERE.
THE WITNESS: THAT’S RIGHT.
THE COURT: AND THE ADVANTAGE FLIP FLOPS WHERE THE RACE
OF THE VICTIM IS BLACK. IN OTHER WORDS, THE ADVANTAGE GOES TD
THE BLACK DEFENDANT WHERE THE VICTIM IS ELACK.
THE WITNESS: OH, YES.
THE COURT: OKAY. I“M READING IT. I WANTED TO MAKE
SURE 1 WAS READING IT RIGHT.
THE WITNESS: YES.
701
BALDUS - DIRECT
1 BY MR. BDGER:
4 @. WHAT CAN YOU CONCLUDE FROM DB-93 AND -4, VIEWED TOGETHER?
A. WELL, IT SUGGESTS A HYPOTHESIS. THE HYPOTHESIS SUGGESTED IS
THAT THE DECISION TO TAKE CASES CUT ON VOLUNTARY MANSLAUGHTER
PLEAS MAY BE AN IMPORTANT POTENTIAL SOURCE OF THE OVERALL RACE
OF VICTIM EFFECT THAT WE SEE, AND THAT THE DECISION TO ADVANCE A
CASE TO A PENALTY TRIAL APPEARS AS A LIKELY CANDIDATE TO EXPLAIN
MUCH OF THE RACE OF VICTIM EFFECT, AND POSSIBLY THE RACE OF
B
R
BE
S
E
E
E
DEFENDANT EFFECT WITHIN WHITE VICTIM CASES, AND THAT THE
10 UNADJUSTED NUMBER SUGGESTS THE HYPOTHESIS THAT THE RACE OF THE
11 VICTIM MAY HAVE SOME EFFECT IN THE DEATH SENTENCING DECISIONS AT
12 PENALTY TRIALS. IT DOES NOT SUPPORT MUCH CONFIDENCE THAT ONE 13 WOULD FIND A RACE OF DEFENDANT EFFECT AT THAT STAGE, HOWEVER.
14 | MR. BOGER: YOUR HONOR, AT THIS TIME I WOULD MOVE THE
13 ADMISSION OF DB-93 AND -94 INTO EVIDENCE.
16 MS. WESTMORELAND: YOUR HONOR, I WOULD ADD AN
17 = ADDITIONAL OBJECTION TQ THESE 2 EXHIBITS. IT SEEMS TO ME THESE
is GET BACK TQ THE PROBLEM OF SOME OF THE EARLIER EXHIBITS, IN NOT
Ww 15 ACCOUNTING FOR ANYTHING EXCEPT FOR THE SENTENCING OUTCOME. AS
20 RACE OF THE DEFENDANT AND RACE OF VICTIM. THAT MAKES THESE
21 DOCUMENTS EVEN LESS RELEVANT THAN SOME OF THE INTERVENING
22 DOCUMENTS.
23 | MR. BOGER: YOUR HONOR, MY RECOLLECTION OF THE SMITH
=4 VERSUS BALKCOM FOOTNOTE IS THAT THE FIFTH AND ELEVENTH CIRCUITS
23 INDICATED INTEREST NOT ONLY IN THE POWER OF POSSIBLE RACIAL
BALDUS ~ DIRECT
1 FACTORS CONTROLLING FOR A WIDE VARIETY OF OTHER FACTORS. BUT
2 INTEREST IN HOW THE CASES WERE DISPOSED OF AS THEY MOVED THROUGH
3 THE SYSTEM FROM INDICTMENT TO GUILTY PLEA. TO ADVANCEMENT TO
4 GUILT TRIAL AND PENALTY TRIAL.
hv’ 1 WOULD SUGGEST THAT DB-93 AND 94 DO PRECISELY THAT.
a 6b as FROFESSOR BALDUS MADE IT IN CLEAR IN HIS TESTIMONY, THERE ARE
7 NO CONTROLS AT THIS POINT, BUT THE FLOW OF CASES THROUGH THE
38 SYSTEM AND THE POSSIBILITY THAT THE FLOW ITSELF WOULD FULLY
? ACCOUNT FOR RACIAL VARIABLES IS SOMETHING THAT THESE TABLES DO
10 ADDRESS IS, I SUBMIT IS RELEVANT.
11 THE COURT: I WILL ADMIT IT, AND THE REASON THAT I WILL
12 ADMIT IT IS PRIMARILY FOR WHAT IS SHOWN ON ROW &. I THINK IT
13 HAS LITTLE IF ANY PROBATIVE WEIGHT OF ANY HYPOTHESIS AS TO WHAT
14 IS PRESENTED PARTICULARLY IN 1, 2, 3, 4, BUT .IT IS A BASIS FROM
13 WHICH I EXPECT. BASED ON THE PATTERN YOU-VE ESTABLISHED, THAT HE
146 DECIDED TO GO A STEP FURTHER. AND FOR PROCESS PURPOSES I WILL
17 LET IT COME IN IN TOTO. AND FOR EVIDENCE ON THE FROOF OF THE
18 MATTERS ASSERTED IN LINE &, I WILL LET IT IN.
Ww 19 MR. BOGER: WELL, I HAD MOVED ALSO 93. AS WELL. YOUR
20 HONOR. I THANK YOU FOR YOUR RULING ON 94, BUT --
21 THE COURT: EITHER %3 DOESN’T FROVE ANYTHING OF ANY
<2 SIGNIFICANCE OR I DON’T YET UNDERSTAND IT. I DON“T KNOW WHICH.
23 MR. BOGER: EITHER WOULD TROUBLE ME, YOUR HONOR. BUT ——
24 I BELIEVE WHAT PROFESSOR BALDUS WAS DOING AT THAT POINT WAS
23 SIMPLY TRYING TO SKETCH OUT VERY QUICKLY IN A SIMPLE TABLE THE
| 203
BALDUS ~ DIRECT
OVERALL RATE AT WHICH WHITE VICTIM CASES AS OPPOSED TO BLACK
VICTIM CASES PROCEEDED THROUGH THE SYSTEM. AND THE OVERALL RATE
{oR PROPORTIONS AT WHICH BLACK DEFENDANT-WHITE VICTIM CASES
STAYED WITHIN THE SYSTEM.
MY SUGGESTION IS THAT HE. THE TABLE REFLECTS. FOR
EXAMPLE. THAT IN THAT COLUMN C, WHAT YOUVE GOT ARE BLACK
DEFENDANT-WHITE VICTIM CASES AT THE OUTSET OF THE SYSTEM
CONSTITUTING ROUGHLY 10 PERCENT OF THE TOTAL POOL OF CASES
INDICTED. AND THAT THEY WIND UP BEING ROUGHLY 40 PERCENT DOWN IN
ROW 7. SO WHAT WE SEE IS THAT WHERE THEY ARE A SMALL GROUP AT
THE START. AT THE WINNOWING OUT PROCESS CONTINUES THROUGH THE
SYSTEM, THEIR PROPORTION RELATIVE TO THE WHOLE INCREASES BECAUSE
THEY REMAIN IN THE SYSTEM DISPROPORTIONATELY WHERE OTHER CASES
ARE DROPPED OUT. AND THE SAME IS TRUE WITH RESPECT TO THE WHITE
VICTIM CASES ANALYZED IN COLUMN B.
THE COURT: MR. BOGER, I WILL ADMIT IT TQ THE EXTENT
THAT IT SETS THE FOUNDATION FOR A NEXT STEP. I DONT THINK IT
DEMONSTRATES ANY FACT OF CONSEQUENCE IN THE CASE WITH SUFFICIENT
RELIABILITY TQ ADMIT IT. |
MR. BOGER: RIGHT, YOUR HONOR.
BY MR. BOGER:
PD. NOW, PROFESSOR BALDUS, THE JUDGE, I THINK, HAS ANTICIPATED
THAT YOU MIGHT WELL HAVE DONE SOME FURTHER ANALYSES THAT
REFLECTED MORE PARTICULARLY ON THE PROSECUTORIAL DECISIONS TO
MOVE A CASE TO THE DEATH PENALTY STAGE AND THE JURY DECISIONS
04
BALDUS - DIRECT
i
|
|
!
|
| THEN OF WHETHER OR NOT TO IMPOSE DEATH.
kJ)
i 1D LIKE TO DIRECT YOUR ATTENTION TO WHATS BEEN MARKED
|
'AS DB-95 FOR IDENTIFICATION ——
S
S
MR. BOGER: -— AND AT THE SAME TIME INFORM THE COURT
A THAT IN DUR REVIEW LAST EVENING OF DB-95., A TYPOGRAPHICAL ERROR
WAS NOTED.
I HAVE SUBSTITUTE PAGES OF BOTH DB-93 AND DB-%6 WHICH 1
&
7
a WOULD LIKE TO SUBSTITUTE AT THIS TIME, IF I MIGHT.
9 I'LL GIVE THE COURT REPORTER A COPY FOR THE RECORD AND
0 A COPY FOR THE COURT.
11 THE COURT: FINE. THANK YOU. THAT IS -935 AND -94&7
}2 MR. BOGER: THAT'S RIGHT, YOUR HONOR.
13 : THE COURT: HAVE YOU SEEN THE NEW ONE, MS.
14 WESTMORELAND?
hy MS. WESTMORELAND: NO. I HAVEN‘T.
16 MR. BOGER: I-M GIVING COPIES TO COUNSEL FOR THE STATE
17 [RIGHT NOW.
13 BY MR. BOGER:
w 19 Q. PROFESSOR BALDUS, DO YOU HAVE DB-95 AS AMENDED?
20 THE COURT: WAIT JUST A MINUTE.
21 MR. BOGER: FORGIVE ME.
22 THE COURT: IF I MOVE MY HAND, MY WHOLE BOOK COMES
23 APART IN MY LAP.
24 WHAT IS 957 IS THAT TABLE 4467
23 THE WITNESS: YES. YOUR HONOR.
203
BALDUS - DIRECT
1 | THE COURT: ALL RIGHT. GQ AHEAD TO -- WAIT A MINUTE.
Zz I HOPE I PUT THE RIGHT ONES BACK IN.
3 WHERE I3 THE TYPOGRAPHICAL ERROR?
4 THE WITNESS: ITS ON DB-9S. IS THAT THE ONE YOU --
4]
THE COULRT: TABLE 447? WHERE IS IT?
& THE WITNESS: ITS OVER IN COLUMN E., IF IT SAYS “IF
7 STATISTICALLY SIGNIFICANT AT .10," TYPED IN, YOU‘VE GOT THE
8 RIGHT ONE.
4 MR. BOGER: YOUR HONOR. ILL BE HAPPY TD HAVE SOMEONE
10 TAKE IT AND ~-
11 THE COURT: THAT'S ALL RIGHT. ILL DO IT AT —-
12 THE WITNESS: THAT .10, YOUR HONOR, THATS THE ONLY
13 TYPO ON THAT PAGE.
14 THE COURT: ILL JUST MAKE THAT BY HAND.
15 oKAY. GO AHEAD.
1&6 BY MR. BOGER:
17 @. PROFESSOR BALDUS, DO YOU HAVE THE AMENDED COPY OF DB-937
1a A. YES.
he 19 QR. CAN YOU IDENTIFY THE DOCUMENT?
20 A. YES. DB-95 IS TABLE 44 FROM OUR REPORT AND IT REPORTS THE
|
21 RESULTS OF MULTIPLE REGRESSION ANALYSES. IN WHICH WE CONTROL FOR
|
22 VARIETY OF BACKGROUND FACTORS.
23 | THESE ANALYSES FOCUS ON THE FPROSECUTORIAL DECISION TO
<4 ADVANCE THE CASE TO A PENALTY TRIAL IN CASES WHERE A CONVICTION
25 OF MURDER HAS BEEN OBTAINED AT A GUILT TRIAL.
a
]
P06
BALDUS - DIRECT
1 THIS IS THE FIRST TABLE THAT WE HAVE PRESENTED THAT
< INCLUDES THE RESULTS FROM BOTH THE CHARGING AND SENTENCING, AND
3 PROCEDURAL REFORM STUDIES.
4 WE INCLUDE THE RESULTS FROM OUR ANALYSES IN THE
S PROCEDIJRAL REFORM STUDY HERE, BECAUSE THAT STUDY ALLOWS US TO
& |FOCUS ON THESE LAST TWO STAGES OF THE CHARGING AND SENTENCING
7 |PROCESS.
8 PART 1 OF DB-9S PRESENTS THE RESULTS OBTAINED IN THE
9 |ANALYSES IN EACH OF THESE STUDIES WITH RESPECT TO THE
10 PROSECUTORIAL DECISION TO ADVANCE THE CASE TO A PENALTY TRIAL.
11 PART A OF SECTION 1 INDICATES THE RACE OF VICTIM
12 EFFECTS ESTIMATED IN THE ANALYSES, WHEREAS PART B PRESENTS THE
13 RACE OF DEFENDANT EFFECTS ESTIMATED,
14 WE STARTED WITH A RELATIVELY SMALL NUMBER OF BACKGROUND
15 CONTROL VARIABLES. AND THEN PROCEEDED TQ INCLUDE MORE VARIABLES.
14 THE RESULTS FOR THE SMALLER ANALYSIS ARE FRESENTED IN
37 COLUMN B, AND THEY INDICATE WITH RESPECT TO THE RACE OF VICTIM,
13 THAT IS, IN ROW ROMAN NUMERAL IA, THAT IN THE CHARGING AND
Ww 19 SENTENCING STUDY, AFTER SIMULTANEOUS ADJUSTMENT FOR ALL THE
20 STATUTORY AGGRAVATING FACTORS. THAT WHITE VICTIMS HAD A 24
=1 PERCENTAGE POINT HIGHER CHANCE, CORRECTION, WHITE VICTIM CASES
22 HAD A 24 PERCENTAGE POINT HIGHER CHANCE OF ADVANCING TO A
23 PENALTY TRIAL THAN DO DEFENDANTS WITH BLACK VICTIMS.
24 THE SECOND ROW LABELED “PROCEDURAL REFORM STUDY" SHOWS
Piel THE CO-EFFICIENT ESTIMATED IN AN IDENTICAL ANALYSIS IN THAT
| 207
BALDUS - DIRECT
1 STUDY.
2 THE PRINCIPAL DIFFERENCES IN THE DATA INVOLVED IN THESE
3 | TWD ANALYSES ARE THAT THE CHARGING AND SENTENCING STUDY HAS
4 |APPROXIMATELY FIFTY MORE PENALTY TRIAL CASES IN THE ANALYSIS,
S AND THE ADDITIONAL VARIABLES THAT WERE NOT AVAILABLE IN THE
6 PROCEDURAL REFORM STUDY.
7 MOVING ACROSS THE PAGE TO COLUMN C. WE SEE THE RESULTS
8 OBTAINED WHEN WE CONTROL FOR ALL OF THE STATUTORY AGGRAVATING
® FACTORS, AS WELL AS A LARGE NUMBER OF MITIGATING FACTORS.
10 THESE ARE THE SAME VARIABLES THAT WE USED IN THE
11 ANALYSIS THAT WAS DISCUSSED YESTERDAY, OF STATUTORY AGGRAVATING
12 FACTORS AND A SUBSTANTIAL PORTION OF MITIGATING FACTORS IN THE
13 |FILE. AND HERE YOU CAN SEE THAT IN THE CHARGING AND SENTENCING
14 STUDY, THE RACE EFFECTS AT THIS STAGE ARE NOT AFFECTED. THEY
15 SLIGHTLY INCREASED.
16 IN THE PROCEDURAL REFORM STUDY. THEY’RE SUBSTANTIALLY
17 |THE SAME AS WELL: A DECLINE OF ONE OR TWO FERCENTAGE POINTS.
13 FINALLY, WHEN WE MOVE OVER TO D AND E, WE SEE THAT IN
w 19 COLUMN D, THE RESULTS WHEN ALL THE FACTORS IN THE FILE WERE 20 INCLUDED AS SIMULTANEOUS BACKGROUND CONTROLS -- THAT IS, IN
|
21 COLUMN D, WE CONTROLLED SIMULTANEOUSLY FOR SOME 230 VARIABLES
5
22 AND IN THE PROCEDURAL REFORM STUDY. WE INTRODUCED BACKGROUND
<3 CONTROLS FOR OVER APPROXIMATELY TWO HUNDRED BACKGROUND
24 VARIABLES, ALL SIMULTANEOUSLY CONTROLLED IN ONE ANALYSIS. AND
23 ONE CAN SEE THAT THE INTRODUCTION OF THOSE ADDITIONAL BACKGROUND
(&
rJ
N
y
o
N
a
— p————— ———— ——————— ———_ —_. — » - = bn
08
BALDUS - DIRECT
CONTROLS REDUCES THE RACE EFFECT ASSOCIATED WITH THE VICTIM, BUT
THAT IT STILL IS SUBSTANTIAL AND STRONG.
FINALLY,» IN COLUMN E, WE SEE THE RESULTS OBTAINED WHEN
WE PRODUCED A SLIGHTLY MORE PARSIMONICUS MODEL THAT WAS
GENERATED THROUGH A STEPWISE REGRESSION ANALYSIS. LIMITING THE
BACKGROUND FACTORS OF THOSE NON-RACIAL VARIABLES THAT WERE
STATISTICALLY SIGNIFICANT AT THE .10 LEVEL, AND WE SEE THERE
RACE EFFECTS PERSIST STRONGLY WITH A HIGH LEVEL OF STATISTICAL
SIGNIFICANCE.
THE STRIKING THING ABOUT THESE ANALYSES, I BELIEVE, IS
THAT THE PATTERN ACROSS ALL THESE ANALYSES IS BASICALLY THE SAME
IN THE TWO STUDIES. THE MAGNITUDE OF THE EFFECTS IS LARGER IN
THE CHARGING AND SENTENCING STUDY, BUT THE CONSISTENCY ACROSS
THESE ANALYSES WHICH ARE BASED ON ENTIRELY DIFFERENT DATA SETS
DEVELOFED WITH DIFFERENT INSTRUMENTS FROM DIFFERENT SOURCES OF
DATA, IS STRIKING EVIDENCE IN MY OPINION OF THE EFFECT THAT RACE
OF VICTIM HAS AT THIS STAGE IN THE CHARGING AND SENTENCING
PROCESS.
NOW. YOUR HONOR, --—
BY MR. BOGER:
@. NOW, THIS STAGE IS THE STAGE WHERE THE PROSECUTOR DECIDES
WHETHER OR NOT TO GO ONTO A PENALTY PHASE AFTER A CONVICTION?
A. YES.
IM NOT SUGGESTING THAT THAT DECISION IS ALWAYS MADE BY
THE PROSECUTOR ALONE. ALL IM SAYING, WHATEVER IS INVOLVED IN
fo
ie
[
ow
R
h
E
t
S
E
R
S
S
G
E
R
fe
J
B
E
o
T
|
f
i
>
rn
0%
BALDUS - DIRECT
MAKING THAT DECISION, BECAUSE SOME OF OUR CORRESPONDENCE WITH
ATTORNEYS SUGGEST THAT IN, AND SUGGESTIONS HAVE BEEN MADE BY
PROSECUTORS IN CORRESPONDENCE, THAT IN SOME OCCASIONS THE JUDGE
PLAYS A ROLE IN THAT DECISION AS WELL. IT IS NOT STRICTLY A
PROSECUTORIAL DECISION. BUT IT’S PLAIN THAT THE PROSECUTOR MUST
INITIATE THE ACTION IN THE FIRST PLACE.
THE COURT: YOU MEAN SOMETHING LIKE, “MR. DISTRICT
ATTORNEY, YOU DON‘T SERIOUSLY MEAN TO TELL ME THAT YOU-RE GOING
TO ASK FOR THE DEATH PENALTY IN THIS CASE." OR --
THE WITNESS: THAT'S PRECISELY ~--
THE COURT: ~—— SOMETHING THAT THAT?
THE WITNESS: EXACTLY. THE RECORD WILL SHOW THAT A
JUDGMENT OF GUILT WAS ENTERED, BY THE JURY, AND THE JUDGE
ENTERED A LIFE SENTENCE, PERIOD, END OF REPORT. NO MORE IN THE
FILE THAN THAT. YOUR HONOR.
SO IT MAY, ITS NOT STRICTLY. I THINK, A PROSECUTORIAL
DECISION.
THE ONLY EVIDENCE WE HAD IS WHETHER OR NOT THERE WAS A
PENALTY TRIAL.
BY MR. BOGER:
|
QR. DID YOU CONDUCT A SIMILAR ANALYSIS. SAY, FOR THE RACE OF THE
DEFENDANT AS WELL?
A. YES, WE DID. AND THOSE RESULTS ARE PRESENTED IN PART i,
SUB-B, SUB—~1 AND SUB-2 OF DB-9% AND THEY SHOW IN THE CHARGING
AND SENTENCING STUDY SUBSTANTIAL RACE OF DEFENDANT DISPARITIES
P10
BALDUS ~ DIRECT
i REGARDLESS OF THE BACKGROUND FACTORS CONTROLLED FOR.
2 THE PROCEDURAL REFORM STUDY. IN CONTRAST, SHOWS MUCH
3 SMALLER RACIAL EFFECTS ASSOCIATED WITH RACE OF DEFENDANT. NONE
4 OF WHICH ACHIEVED A LEVEL OF STATISTICAL SIGNIFICANCE.
3 I MIGHT ADD IN CONNECTION WITH THE PROCEDURAL REFORM
STUDY DATA HERE, YOUR HONOR, THAT THOSE, THAT THAT ANALYSIS
INCLUDED ALL THE DATA SO THAT. THAT IN ALL THE DATA CONCERNING
PENALTY TRIALS, AND THOSE MEASURES OF STATISTICAL SIGNIFICANCE.
B
E
G
+
S
L
PR
GIVE US SOME ESTIMATE. HOWEVER. OF THE LIKELIHOOD THAT THE
10 |RANDOM CHANCE FACTORS OPERATING IN THE SYSTEM MIGHT BE THE
11 SOURCE OF THOSE DISPARITIES.
12 AGAIN, I THINK ITS IMPORTANT TO EXAMINE THE PATTERN
13 [THAT ALL OF THESE CO-EFFICIENTS RUN IN THE SAME DIRECTION. AND
14 THAT THEY PERSIST ACROSS EACH SEPARATE ANALYSIS, SUGGESTING THAT
15 THE RACE OF DEFENDANT DOES HAVE AN EFFECT AT THIS STAGE, BUT
16 AGAIN BECAUSE THE EFFECTS ARE NOT SYSTEMATIC AND STRONG ACROSS
17 EVERY ANALYSIS. THE EVIDENCE DOES NOT SUGGEST THEY ARE NEARLY AS
18 STRONG OR PERSISTENT AS WE OBSERVED WITH RESPECT TO THE RACE OF
w 19 VICTIM.
20 THE COURT: AND UNLIKE THOSE FOR THE VICTIM, I NOTICED
21 YOUR STATISTICAL SIGNIFICANCE LEVELS ARE NOT GENERALLY AS GOOD.
22 THE WITNESS: THATS TRUE, YOUR HONOR. THAT’S QUITE
23 RIGHT, ESPECIALLY WITH RESPECT TO THE PROCEDURAL REFORM STUDY,
24 THEY DO NOT APPROACH .0S5 LEVEL OF SIGNIFICANCE.
23 THE. THE ONLY. AGAIN I EMPHASIZE WE HAVE HERE ALL THE
fp
Gg
O
N
>
G
b
W
N
F111
BALDUS - DIRECT
CASES. THE QUESTION IS WHETHER THERE-“S AN EFFECT OUT THERE IS.
IS IN PART ANSWERED BY THE FACT THAT YOU GET THESE SAME, THE
SAME DIRECTION OF THOSE CO-EFFICIENTS., THAT IS. THAT EACH ONE IS
IN A POSITIVE DIRECTION. IF THERE WERE NO EFFECTS OUT THERE, IT
WOULD BE LIKELY YDU WOULD SEE THESE CO-EFFICIENTS BOUNCING
ABOUT, SOME POSITIVE AND SOME NEGATIVE. THAT” S SOME EFFECT. BUT
AS WE HAVE EMPHASIZED THROUGHOUT. THE EVIDENCE ON THE RACE OF
THE DEFENDANT IN THIS SYSTEM, WHEN YOU LOOK AT IT AS A WHOLE, IS
MUCH LESS SYSTEMATIC, MUCH LESS STRONG THAN IT IS WITH RESPECT
TO THE RACE OF THE VICTIMS. THAT’S WHAT THE NUMBERS TELL US.
THOSE ARE THE CONCLUSIONS THAT ARE SUPPORTED BY THESE DATA.
NOW WHEN WE LOOK DOWN AT THE JURY SENTENCING DECISIONS
WE SEE A DIFFERENT PATTERN. WE SEE WITH RESPECT TO THE RACE OF
THE VICTIM THAT IN BOTH THE CHARGING AND SENTENCING STUDY. AND
THE PROCEDURAL REFORM STUDY, THAT THERE ARE RACE OF VICTIM
EFFECTS IDENTIFIED.
IN THE ANALYSIS INVOLVING ALL THE STATUTORY AGGRAVATING
FACTORS AND IN THE SUBSEQUENT ANALYSIS INVOLVING THE LARGE
NUMBER OF MITIGATING FACTORS, AS WELL.
THESE EFFECTS ARE REDUCED IN TERMS OF THEIR LEVEL OF
STATISTICAL SIGNIFICANCE WHEN WE EXAMINE ALL OF THE FACTORS IN
THE FILE. WE DID THIS BY WAY OF A STEPWISE REGRESSION, BECAUSE
THE NUMBER OF BACKGROUND FACTORS YOUR HONOR. IS SO LARGE THEY
WOULD COMPLETELY QUVERWHELM AN ANALYSIS WHERE THEY WERE ALL
INCLUDED IN THE ANALYSIS AT ONE TIME. AND THE NUMBER OF
M
O
W
o
BALIUS - DIRECT
BACKGROUND FACTORS CLOSELY APPROXIMATE THE NUMBER OF CASES IN
EACH OF THE SAMPLES, IN FACT, IT EXCEEDS THE NUMBER OF CASES IN
THE PROCEDURAL REFORM STUDY SAMPLE. THEREFORE. FOR TECHNICAL
REASONS. YOU CAN‘T CONTROL FOR ALL OF THEM ALL SIMULTANEOUSLY.
AGAIN, I THINK THE EVIDENCE HERE SUGGESTS THAT THERE
ARE RACE OF VICTIM EFFECTS AT THIS LEVEL. THEY ARE NOT NEARLY
AS STRONG OR PERSISTENT AS WHAT WE SEE AT THIS LEVEL, WHERE THE
DECISIONS ARE TAKEN TO ADVANCE A CASE TO A PENALTY TRIAL.
FINALLY, WE LOOK DOWN AT THE RACE OF DEFENDANT EFFECTS.
AND HERE YOU SEE THE EXAMPLE OF A SYSTEM WHERE THERE APPEAR TO
BE NO SYSTEMATIC EFFECTS AT ALL.
YOU“LL NOTICE THAT THE CO-EFFICIENTS BOUNCE AROUND FROM
POSITIVE TO NEGATIVE, IN FACT MOST OF THEM APPEAR ON THE
NEGATIVE SIDE. AND NONE OF THEM EVEN BEGINS TO APPROACH A HIGH
LEVEL OF STATISTICAL SIGNIFICANCE, OR EVEN A MODEST LEVEL, AND
THESE ARE THE KINDS OF RESULTS THAT WOULD CLEARLY LEAD ONE TO
CONCLUDE THAT THESE DISPARITIES YOU SEE HERE ARE VERY LIKELY THE
RESULT OF CHANCE FACTORS OPERATING IN THAT SYSTEM, RATHER THAN
|THE RESULT OF ANY REAL EFFECT IN THE SYSTEM. AND IF THERE IS
ANY EFFECT, IT’S A VERY MODEST ONE.
BY MR. BOGER:
RA. PROFESSOR BALDUS, DO YOUR REMARKS WITH RESPECT TO THE JURY
SENTENCING DECISION ON RACE OF DEFENDANT REFLECT THAT THERE IS
NO RACIAL DISPARITY ON THE, ON DEFENDANT PARAMETER ANYWHERE IN
THE SYSTEM OR ONLY AT THIS DECISION POINT?
13
BALDUS ~ DIRECT
1 A. NO. THIS ANALYSIS FOCUSES ONLY ON BEHAVIOR OF THE JURORS
2 WHO HAVE AN IMPORTANT EFFECT TO BE SURE IN THE PROCESS, BUT THEY
3 OBVIOUSLY ARE NOT THE SOURCE OF THE RACIAL DISPARITY THAT WE SEE
WHEN WE LOOK AT THE COMBINED EFFECTS OF ALL THE DECISIONS IN THE
PROCESS.
THE ANALYSIS THAT WE LOOKED AT FOR THE LAST DAY SHOW.
REFLECTS THE DISPARITIES AND THE LIKELIHOOD THAT BLACK OFFENDERS +-
PARDON ME —- THAT THE LIKELIHOOD THAT OFFENDERS OF BLACK AND
Co
PE
RC
B
E
E
TN
e
t
ER
WHITE VICTIMS INDICTED FOR MURDER WILL END UP RECEIVING A DEATH
10 SENTENCE.
i1 IT APPEARS FROM THIS THAT THE JURY. THE JURIES IN THE
12 PENALTY TRIALS HAVE A, TAKE RACE OF THE VICTIM INTO ACCOUNT TO A
13 CERTAIN EXTENT. THATS WHAT‘S SUGGESTED BY 2 SUB A. BUT WITH
14 SPECT TO THE DEFENDANT, IF ANYTHING, THEY HAVE A CORRECTIVE
13 EFFECT IN TERMS OF ANY ADVERSE EFFECT OF EARLIER DECISIONS ON
14 BLACK DEFENDANTS. IF ANYTHING, THESE DATA CONCERNING THE BLACK
37 DEFENDANT AT A PENALTY TRIAL SUGGEST THAT THE BLACK DEFENDANTS
18 |MAY BE SLIGHTLY BETTER OFF, BUT THE DISPARITIES ARE SO SMALL
wo 1® THAT ONE CANT DRAW ANYTHING FROM THEM REALLY.
20 THE COURT: AND THE SIGNIFICANCE OF ALL THE FIGURES IN
21 “Ev COLUMN IS NOT AS GDOD AS ONE WOULD HOPE FOR.
22 THE WITNESS: WELL, ——
23 : THE COURT: NOT NEARLY SO SIGNIFICANT AS SOME OF THE
24 OTHER FIGURES YOU‘VE BEEN ABLE TO PRODUCE --
2% THE WITNESS: THAT'S RIGHT.
214
BALDUS - DIRECT
1 | THE COURT: --— AT THE JURY LEVEL.
THE WITNESS: RIGHT, AT THE JURY LEVEL. THE EFFECTS ARE
MUCH SMALLER. VYOU’RE TALKING ABOUT A FIVE PERCENTAGE POINT
2
3
4 DIFFERENCE ON AVERAGE IN THESE FIGURES.
S AND AGAIN I EMPHASIZE THAT THESE DISPARITIES REFLECT
& THE AVERAGE EFFECT ACROSS ALL OF THE CASES, AND THE ANALYSES
7 THAT WE“VE DONE WITH THE JURIES IN WHICH WE BREAK THEM OUT
a ACCORDING TO HOW SERIOUS THEY ARE, HOW SERICUS THOSE CASES ARE.
? SUGGEST THAT HERE ALSO, THAT THESE EFFECTS TEND TO BE
10 CONCENTRATED IN THE MORE AMBIGUOUS.
33 IN THE MID-RANGE OF CASES WHERE IT-S NOT SO CLEAR, IN
12 JURY DECISION MAKING AS WELL, WITH RESPECT TO THE RACE OF THE
13 VICTIM, WHEN THE CASES ARE REALLY HIGHLY AGGRAVATED, THERE IS
14 ABSOLUTELY NO SIGN OF RACE OF THE VICTIM EFFECT. WHEN THEY RE
13 NOT AT ALL NOT AGGRAVATED, THERE’S NO SIGN OF THE RACE OF THE
1&6 VICTIM EFFECT.
17 WHAT YOU’RE SEEING IN HERE ARE THE AVERAGE EFFECTS OF
183 THOSE RELATIVELY SMALL POCKET OF CASES IN THE MIDDLE, AMONG ALL
FS 19 THE CASES THAT COME BEFORE JURIES, WHERE THERE IS EVIDENCE OF A
20 RACE OF VICTIM EFFECT. THATS WHAT WE SEE THE RESIDUE OF IN
21 THESE OVERALL CD-EFFICIENTS HERE.
za THE COURT: LET ME ASK YOU A QUESTION AT THIS POINT.
23 ARE YOU FAMILIAR WITH ANY LITERATURE THAT WOULD SUGGEST
24 AT AN EARLIER TIME PERIOD WHAT THE "EE" COLUMN MIGHT HAVE LOOKED
25 LIKE UNDER JURY SENTENCING DECISIONS?
oi
e
o
g
FR
EE
~
OH
TE
RR
E
C
J
R
S
E
R
E
713
BALDUS - DIRECT
THE WITNESS: YES, YOUR HONOR. I‘M FAMILIAR WITH THE
LITERATURE, AND I“M ALSO FAMILIAR WITH SOME OF MY OWN ANALYSIS.
THE COURT: WELL. I AM WONDERING IF THE SUPREME COURTS
INSISTENCE ON CONSTITUTIONALLY DRAWN JURIES IS HAVING ANY EFFECT
IN YOUR VIEW ON THE EVENHANDEDNESS OF JURY DECISION MAKING.
THE WITNESS: WELL. IT IS MY OPINION THAT IT HAS HAD AN
EFFECT. I BASE THAT OPINION ON --
THE COURT: A POSITIVE EFFECT?
THE WITNESS: YES. 1 BASE THAT OPINION ON AN ANALYSIS
OF JUST THE FIGURES YOU REFER TO. YOUR HONOR, ON A SAMPLE OF
CASES DRAWN IN THE PRE-FURMAN PERIOD IN THIS STATE. AND IT’S A
MUCH SMALLER SAMPLE TO BE SURE, SO I DON’T WANT TO MAKE TOO MUCH
OF IT. IT“S MUCH SMALLER THAN THESE SAMPLES WE’RE DEALING WITH
HERE. :
BUT IT SHOWS IN TERMS OF JURY SENTENCING, IT SHOWS
QUITE SIGNIFICANT RACE OF DEFENDANT EFFECTS.
INTERESTINGLY, HOWEVER, IT DOESNT SHOW QUITE AS STRONG
RACE OF VICTIM EFFECTS AS SOME OF THESE ANALYSES DO IN THE
POST~-FURMAN PERIOD.
THAT IS, THE DATA SUGGESTS BY THIS EARLIER WORK, WHICH
1s NOT INCLUDED, I MENTION IN PASSING, IN OUR REPORT HERE, BUT
THAT SUGGEST THAT THE RACE (OF DEFENDANT EFFECT IS DIMINISHED IN
JURY SENTENCING IN THIS POST-FURMAN PERIOD, AND I BELIEVE ON THE
BASIS OF MY STUDY OF THE SUBJECT THAT IT IS IN SIGNIFICANT PART
ATTRIBUTED TO THE FACTS YOU ALLUDED TO, PLUS CHANGES IN
BALDUS - DIRECT
1 ATTITUDES. POPULATION. VARIETY OF FACTORS THAT WOULD PRODUCE
2 THAT EFFECT. PERHAPS AS WELL AS SOME IMPACT OF THE STATUTE. I
3 DONT KNOW.
4 BUT THE RACE OF THE VICTIM EFFECTS. HOWEVER, SEEM TO BE
. LARGER THAN THEY WERE IN THE PRE-FURMAN PERIOD. THAT’S THE
il b THING THAT STRUCK ME IN TERMS OF MAKING THAT COMPARISON.
7 THE COURT: THAT IS CURIOUS.
8 MR. BOGER: YOUR HONOR, I MOVE THE ADMISSION OF DB-93
9 INTO EVIDENCE. |
10 THE COURT: I WILL ADMIT IT UNLESS YOU HAVE ANY
11 |PARTICULAR OBJECTION —-
12 MS. WESTMORELAND: NO NEW OBJECTION.
13 THE COURT: -- NEW OBJECTION BEYOND THOSE PREVIOUSLY
14 (STATED.
15 MR. BOGER: THANK YOU.
16 THE COURT: ALL RIGHT, GO AHEAD, MR. BOGER. 95 IS
17 |ADMITTED.
13 MR. BOGER: THANK YOU. YOUR HONOR.
w 19 BY MR. BOGER1
20 |G. NOW, PROFESSOR BALDUS. HAVE YOU EMPLOYED ANY OTHER
21 |STATISTICAL METHODS TO ANALYZE PROSECUTORIAL DECISION MAKING IN
22 |THE CHARGING AND SENTENCING SYSTEM?
23 |A. YES. WE’VE UNDERTAKEN A SERIES OF ANALYSES USING LOGISTIC
24 REGRESSION PROCEDURES.
bh
A Q. LET ME ASK YOU TO TURN TD DB-%& MARKED FOR IDENTIFICATION,
|
|
|
|
i
1
BALDUS - DIRECT
THE AMENDED COPY. AND IF YOU WOULD, STATE THE DIFFERENCE BETWEEN
THE AMENDED COPY OF DB-94 AND THE ORIGINAL THAT WE SUBMITTED
EARLIER IN YOUR EVIDENCE BOOK?
A. YES. THE DIFFERENCE IS THAT IN COLUMN A IN THE FOURTH ROW
BELOW THE LINE, THE TEXT WAS NOT PROPERLY ADJUSTED. THAT SHOULD
BE A “C" WHICH FOLLOWS AFTER "A" AND "B". IT WAS INADVERTENTLY
GIVEN A SEFARATE NUMBER IN THE ORIGINAL OF THIS EXHIBIT..
@. ALL RIGHT, CAN YOU IDENTIFY THE AMENDED VERSION, THEN, OF
DB~967
A. YES. IT IS A PRESENTATION OF THE RESULTS OF THE LOGISTIC
REGRESSION ANALYSES THAT WERE CONDUCTED OF THE DECISION TO
ADVANCE A CASE TO A PENALTY TRIAL WITH BACKGROUND CONTROLS
INTRODUCED, WERE A VARIETY OF LEGITIMATE NON-RACIAL FACTORS.
THE RESULTS ARE FROM SUCH ANALYSES IN BOTH THE CHARGING AND
SENTENCING AND PROCEDURAL REFORM STUDY.
THE DIFFERENCE HERE IS IN THE TYPE OF REGRESSION
ANALYSIS USED, AND THE LOGISTIC APPROACH HAS PARTICULAR
RELEVANCE TO THE ANALYSIS IN THIS PROCEEDING, YOUR HONOR,
BECAUSE THIS IS A PROCEDURE THAT RESTS ON A SOMEWHAT DIFFERENT
ASSUMPTION THAN THE LEAST SQUARES ANALYSES DID.
IT ASSUMES THAT THE EFFECTS OF VARIABLES ARE GOING TO
BE MUCH LIKE THE EFFECTS THAT WE SEE IN CUR ANALYSIS OF THE
DIFFERENT LEVELS OF SERIOUSNESS IN THE CASES. THAT IS, THE
ASSUMPTION OF A LOGISTIC ANALYSIS IS THAT THE EFFECTS ARE GOING
TO BE MORE IMPORTANT IN THE MIDDLE RANGE OF CASES, THAT IS,
918
BALDUS — DIRECT
WHERE THE PROBABILITY OF SELECTION IS NOT AS HIGH. WE‘LL SEE A
GREATER EFFECT. THAT'S A PROPERTY OF THE REGRESSION PROCEDURE.
| WHEREAS, THE LEAST SQUARES REGRESSION ANALYSIS ASSUMES
THAT THE EFFECT IS GOING TO BE THE SAME ACROSS ALL CATEGORIES OF
RISKS PRODUCED BY OTHER FACTORS, THE LOGISTIC PREMISE LOOKS LIKE
A CURVE LIKE THIS, IT RISES IN THE MIDDLE, DROPS AT EITHER END.
AND SUGGESTS THAT THE FACTORS, NOT JUST THE RACIAL FACTORS, BUT
ALL FACTORS WILL HAVE THEIR BIGGEST IMPACT IN THIS CATEGORY OF
CASES. |
SO FOR THAT REASON. IT HAS PARTICULAR APPLICABILITY IN
|ANALYSIS OF DATA OF THE TYPE WE HAVE HERE.
AND YOU CAN SEE IN LOOKING AT DB~96 THAT THE RESULTS OF
THIS LOGISTIC ANALYSIS WITH RESPECT TO THE RACE OF VICTIM SHOW
IN BOTH STUDIES, BOTH STUDIES, THE RESULTS IN BOTH STUDIES ARE
IDENTIFIED IN COLUMN A, CHARGING AND SENTENCING STUDY, AND BELOW
THAT THE PROCEDURAL REFORM STUDY. SHOW VERY SUBSTANTIAL RACE OF
VICTIM CO-EFFICIENTS IN ALL OF THESE ANALYSES, AND HIGH LEVELS
OF STATISTICAL SIGNIFICANCE.
Gl. WHAT PROSECUTORIAL DECISION MAKING FUNCTIONS DOES THIS
ANALYSIS FOCUS UPON?
A. THIS FOCUSES ON THE DECISION TO ADVANCE THE CASE TO A
PENALTY TRIAL.
Q. ALL RIGHT.
A. NOW, COLUMNS D AND E SHOW THE CO-EFFICIENTS IN THE DEATH
ODDS MULTIPLIERS CALCULATED FROM THOSE CO-EFFICIENTS THAT RELATE
212
BALIUS - DIRECT
TO THE RACE OF THE DEFENDANT. AND THEY ARE CONSISTENT WITH THE
RESULTS WE SAW FROM THE LEAST SQUARES ANALYSIS EARLIER. THAT
IS, IF THE CHARGING AND SENTENCING STUDY WE SEE EFFECTS THAT ARE
NOT AS STRONG AS THE, THOSE ESTIMATED IN CONNECTION WITH THE
RACE OF THE VICTIM, BUT THEY REACH LEVELS OF STATISTICAL
SIGNIFICANCE, WHILE IN THE PROCEDURAL REFORM STUDY WE SEE VERY
MODEST EFFECTS WHICH DO NOT ACHIEVE LEVELS OF STATISTICAL
SIGNIFICANCE.
SO THIS ANALYSIS PRESENTS MUCH THE SAME PICTURE THAT WE
saW IN THE PREVIOUS EXHIBIT, DB-95, WHICH USED A DIFFERENT
MULTIPLE REGRESSION PROCEDURE, THAT RESTS ON SOMEWHAT DIFFERENT
TECHNICAL ASSUMPTIONS.
MR. BOGER: YOUR HONOR, I MOVE THE ADMISSION OF DB-%26
INTO EVIDENCE.
MS. WESTMORELAND: SAME OBJECTIONS. YQUR HONOR.
THE COURT: THIS IS FOCUSING ON THE DECISION MAKING AT
THE POINT WHERE IT IS UP TO THE PROSECUTOR. THEORETICALLY. TO
ADVANCE IT TO A PENALTY PHASE?
THE WITNESS: YES. YOUR HONOR.
THE COURT: THE FIVE LEGITIMATE FACTORS ARE THE FIVE
AGGRAVAT ING CIRCUMSTANCES, OR CAN YOU ——
THE WITNESS: NO, NO. THOSE FIVE LEGITIMATE FACTORS,
YOUR HONOR, ARE THE FIVE THAT I REFERRED TO YESTERDAY.
THE COURT: PRIOR FELONY OR SOMETHING.
THE WITNESS: YES, YOUR HONOR.
20
BALDUS - DIRECT
jo
o THE COURT: WILL YOU REFRESH MY RECOLLECTION OF WHAT
2 THEY ARE?
3 THE WITNESS: CERTAINLY. CERTAINLY.
a THE COURT: WELL, YOU DON‘T HAVE IT EITHER, DO YOU?
?
THE WITNESS: WELL, I -—- (&
6 THE COURT: WAIT A MINUTE. WAIT A MINUTE. IT SEEMS TO
7 ME FOR MY PURPOSES THAT ROMAN NUMERAL II. LINE ROMAN NUMERAL II
8 AND LINE ROMAN NUMERAL “D* MIGHT BE THE MOST INTERESTING. SO
? LETS TALK ABOUT THOSE FOR A MINUTE.
0 THE WITNESS: ALL RIGHT, YOUR HONOR.
3s THE COURT: GIVE ME SOME SORT OF AN IDEA OF THIS
12 NON-ARBITRARY FACTOR. THAT IS A TERM THAT I DON’T KNOW THAT
13 WE’VE DISCUSSED. WHAT DOES THAT REFER —
14 THE WITNESS: THAT REFERS. YOUR HONOR. TO THE APFROACH
13 THAT WE USED THAT WAS INSPIRED BY GODFREY V. GEORGIA, WHEREIN WE
1&6 SOUGHT TO LIMIT THE BACKGROUND CONTROLS TO FACTORS THAT WERE
17 FIRST, FROM AN A PRIORI STANDPOINT WERE LEGITIMATE TO CONSIDER.
18 THAT 1S, WOULD NOT INCLUDE SUSPECT FACTORS THAT RELATED TO
W 17 OFFENDERS SEX. INTELLIGENCE, SOCIO-ECONOMIC STATUS, ITS |
20 LIMITED TO JUST TO LEGITIMATE MITIGATING-AGGRAVATING FACTORS.
21 IT ALSO LIMITED THE VARIABLES TO AGGRAVATING FACTORS
22 THAT HAD AN AGGRAVATING EFFECT IN THE ANALYSIS, AND MITIGATING
23 FACTORS THAT HAD A MITIGATING EFFECT IN THE ANALYSIS. IF IT
<4 WERE A VARIABLE THAT WAS PERCEIVED FROM THE STANDPOINT OF LAW TO
23 HAVE A MITIGATING EFFECT, OR PRODUCE AN AGGRAVATING EFFECT. THAT
221
BALIUS ~ DIRECT
WOULD NOT BE PERCEIVED FROM THIS PERSPECTIVE TO BE APPROPRIATE
OR RATIONAL BASIS FOR IDENTIFYING SIMILAR CASES. IT WAS BASED
ON THE JUDGMENT IN GODFREY. THAT THERE ARE CERTAIN CASES THAT
CANNOT RATIONALLY BE USED TD DETERMINE WHAT CASES ARE COMPARABLE
FOR PURPOSES OF EVALUATING ARBITRARINESS AND EVENHANDEDNESS OF
TREATMENT OF CASES IN THE SYSTEM.
MR. BOGER: YOUR HONOR, I HAVE COPIES OF THE TECHNICAL
APPENDIX WHICH WOULD INCLUDE. I THINK, IT LOOKS LIKE SCHEDULE 8.
THE LIST OF FACTORS THAT PROFESSOR BALDUS IS REFERRING TO.
THE COURT: REFERRING TO NON-ARBITRARY FACTORS, QUOTE.
UNQUOTE?
MR. BOGER: THAT‘S CORRECT, YOUR HONOR.
THE COURT: PASS IT UP AND LET ME LOOK AT IT.
MR. BOGER: BE GLAD TO.
THE COURT: IS IT IN ENGLISH?
MR. BOGER: WE WORKED LAST EVENING. I THINK IT IS IN
ENGLISH, YOUR HONOR.
PROFESSOR BALDUS, I BELIEVE, HAS A COPY. AND I“LL GIVE
A COPY TO THE CLERK, AND SUGGEST I‘M NOT GOING TO ADMIT IT,
ITLL COME IN WITH THE FINAL REPORT. BUT FOR THESE PURPOSES IT
CAN BE MARKED AS DB-96A FOR IDENTIFICATION.
THE COURT: WHAT PAGE IS IT?
THE WITNESS: THIS IS ON —-— IN THE APPENDIX, YOUR
HONOR?
THE COURT: WHAT PAGE AM I SUPPOSED TO LOOK AT?
222
BALDUS ~ DIRECT
Fe
y THE WITNESS: SCHEDULE 12, YOUR HONOR.
2 THE COURT: WHAT IS DEFENDANTS STATUS SYMPATHETIC
3 MEASURE?
a THE WITNESS: THAT IS A PERSON WHO IS RETIRED OR A
5 STUDENT, YOUR HONOR. IT HAS A STATISTICAL EFFECT ON THE
4 b BEHAVIOR OF THE PEOPLE IN THE SYSTEM IS THE REASON IT’S
7 INCLUDED.
8 THE COURT: OKAY.
9 I DON’T KNOW, DO YOU INTEND TO OFFER THIS WHOLE PACKAGE
10 YOUVE HANDED ME AS AN EXHIBIT?
11 MR. BOGER: YOUR HONOR, IT‘S PART OF A TECHNICAL
12 APPENDIX FOR PROFESSOR BALDUS” REPORT. WE INTEND TO OFFER THE
13 ENTIRE REFORT.
14 THE COURT: I HAVEN'T SEEN THAT AND I HAVE NO IDEA
15 WHETHER I-LL ADMIT IT OR NOT.
16 MR. BOGER: I RECALL YESTERDAY YOUR HONOR ASKED FOR
17 ENGLISH DEFINITIONS OF SOME 75 FACTORS, AND WE WORKED TO PUT
13 THAT TOGETHER, AND WE DIDN‘T KNOW WHAT ELSE MIGHT COME UP, SO
w 19 WE PUT THE WHOLE PACKAGE BEFORE THE COURT.
20 THE COURT: ALL RIGHT. WELL, THE RECORD OUGHT TO SHOW
- THAT SCHEDULE 12 OF THE TECHNICAL APPENDIX AS TRANSLATED INTO
22 ENGLISH IS NEEDED FOR THE UNDERSTANDING OF DB-946 AND AT LEAST
23 SCHEDULE 12 IS ADMITTED.
24 ALL RIGHT. NOW, THIS BRINGS UP A POINT THAT YOU HAVE
5 MENTIONED SEVERAL TIMES. THIS DOES NOT TAKE INTO ACCOUNT A
BALDUS - DIRECT
1 FACTOR WHICH I WOULD IMAGINE VIRTUALLY EVERYBODY FAMILIAR WITH
2 THE CRIMINAL JUSTICE SYSTEM WOULD SAY WOULD BE HIGHLY OPERATIVE
3 AT THIS LEVEL. AND THAT“S STRENGTH OF THE EVIDENCE.
4 DO WE ADDRESS THAT LATER?
3 THE WITNESS: NO. IN THIS ANALYSIS. YOUR HONOR, WE
4 & HAVE CONTROLLED FOR THE STRENGTH OF THE EVIDENCE BY LIMITING THE
~4
CASES TO THOSE THAT RESULTED IN A MURDER CONVICTION AT A JURY
3 TRIAL. THIS ANALYSIS DOES NOT CONTAIN ANY INDEPENDENT MEASURES
? OF THE STRENGTH OF EVIDENCE. WE DO NOT HAVE SUCH MEASURES IN
10 THE PROCEDURAL REFORM STUDY.
11 | HOWEVER, IN THE --
12 THE COURT: LET ME GIVE YOU A CLASSIC EXAMPLE.
13 THE WITNESS: RIGHT.
14 THE COURT: I DON‘T KNOW OF A MURDER CASE THAT IS MORE
15 EXACERBATED THAN THE MURDER OF THE 20. 20, 40, HOWEVER MANY IT
16 WAS. BLACK CHILDREN WE HAD IN ATLANTA A YEAR OR SO AGO. IF I
17 UNDERSTAND WHAT WENT ON IN THAT CASE. THE DEATH PENALTY WASN-T
18 SOUGHT. AND THE REASON THE DEATH PENALTY WASN'T SOUGHT WAS THAT
w» 19 THEY DIDN‘T THINK THE EVIDENCE WAS STRONG ENOUGH TO ASK FOR IT.
20 | MY ARMCHAIR VIEW IS THAT THAT OPERATES AN AWFUL LOT ON
|
21 THE PROSECUTORS DECISION MAKING. AND MY QUESTION TO YOU SIMPLY
22 (IS, BEYOND WHAT YOQU’VE JUST SAID TO ME. HAVE YOU USED ANY OF THE
23 STRENGTH OF EVIDENCE DATA THAT YOU DID COLLECT TO TRY TO LOOK AT
24 IT AT THIS LEVEL?
ra
n
THE WITNESS: YOUR HONOR, -——
BALDUS - DIRECT
1 THE COURT: I KNOW YOU DIDN’T WITH THIS, THAT'S CLEAR,
2 BUT HAVE YOU IN A SUBSEQUENT --—
THE WITNESS: ACTUALLY THE PRECEDING ANALYSIS THAT WE
4 EXAMINED USING THE LEAST SQUARES ANALYSIS. YOUR HONOR, DID
3 INCLUDE CONTROLS FOR ALL OF DUR EVIDENTIARY FACTORS. THOSE WERE
4 6 INCLUDED = . :
7 THE COURT: TO THE EXTENT THEY WERE STATISTICALLY
8 SIGNIFICANT.
? THE WITNESS: NO, NO, YOUR HONOR. IN DB-95, --
10 THE COURT: IN COLUMN E?
11 THE WITNESS: WELL. ACTUALLY, YOUR HONOR. IN COLUMN D.
12 IT INCLUDED THEM ALL. WHETHER THEY WERE STATISTICALLY
13 SIGNIFICANT OR NOT. THE ENTIRE FILE OF EVIDENTIARY VARIABLES
13 THAT WE HAD —-.
13 THE COURT: WHAT 1“M SUGGESTING TO YOU IS THAT ON THAT
16 SORT OF BASIS IS IT WEIGHED EACH AND EVERY FACTOR EQUALLY.
17 THE WITNESS: EACH AND EVERY FACTOR WENT INTO THE
18 ANALYSIS. THEN THE ANALYSIS DETERMINES, ESTIMATES THE
w 19 CO-EFFICIENT FOR EACH OF THOSE EVIDENTIARY FACTORS. AND GAVE IT |
20 A WEIGHT, DEPENDING UPON ITS IMPORTANCE IN PREDICTING THE
21 DECISION TO SEEK A DEATH PENALTY.
22 THIS ANALYSIS THATS REPORTED HERE IN COLUMN D., YOUR
23 HONOR, DB-95, ADJUSTS SIMULTANEOUSLY FOR ALL THE AGGRAVATING AND
24 MITIGATING FACTORS AS WELL AS ALL THE EVIDENTIARY FACTORS,
23 EVIDENTIARY VARIABLES THAT ARE IN OUR ANALYSIS. THEY ARE
225
BALDUS - DIRECT
1 INCLUDED IN THIS ANALYSIS HERE.
2 THE COURT: I UNDERSTAND THAT.
3 BUT IF I UNDERSTAND WHAT YOU HAVE TOLD ME, JUST SEIZING
ON THE WILLIAM WILLIAMS CASE AS AN EXAMPLE, MAYBE YOU CAN a
3 RESPOND TO IT. I WOULD IMAGINE, AND LET”S JUST TAKE IT AS TRUE,
& & THAT THE ONLY REASON THE DEATH PENALTY WASN‘T SOUGHT IN THAT
iT CASE WAS THE STRENGTH OF THE EVIDENCE.
8 AT THAT POINT IN TIME THAT BECOMES MORE EQUAL THAN ALL
@ THE OTHER VARIABLES.
10 NOW, I HAVE AN OPINION THAT THAT IS PROBABLY MORE
11 IMPORTANT THAN ANY OTHER SINGLE VARIABLE ONCE YOU HAVE GOTTEN A
12 MURDER CONVICTION,
13 WILL WHAT YOU DID IN COLUMN D ON DB-93 TAKE INTO
14 ACCOUNT THE FACT THAT THAT FACTOR MAY BECOME MORE IMPORTANT THAN
15 ANY OTHER FACTOR, AT LEAST CONSCIOUSLY MAY BECOME MORE IMPORTANT
1&6 THAN ANY OTHER FACTOR AT THAT LEVEL, AND GIVE DUE CONSIDERATION
17 TO IT. IN OTHER WORDS, IF I REMEMBER YOUR HALOTHANE ANALYSIS,
1a YOU SHOULD HAVE COME UP WITH THESE FIGURES BASED ON LITTLE
w 12 CLUSTERS OF SIMILAR CASES WHERE EVERYTHING WAS SIMILAR,
20 INCLUDING STRENGTH OF THE EVIDENCE?
21 : THE WITNESS: THAT’S RIGHT. YOUR HONOR.
22 THE COURT: MY QUESTION TQ YOU IS, DO YOU THINK YOU
23 HAVE ADJUSTED IN "D" FOR THIS KIND OF VARIABLE WHICH MAY BE S0
24 SUBSUMING OF ALL OTHER VARIABLES? HAS THAT BEEN GIVEN ADEQUATE
29 ATTENTION IN “D" IN YOUR VIEW OF THE STATISTICAL TECHNIWUES
+:
ES,
T
RE
1 J
ER
E
0
BALDUS ~ DIRECT
INVOLVED?
THE WITNESS: I BELIEVE, IT HAS, YOUR HONOR, IN THE
SENSE WE HAVE A VARIETY OF VARIABLES THAT MEASURE EXACTLY THE
KIND OF VARIABLE RELATING TO EVIDENCE THAT WERE INVOLVED IN THE
WILLIAMS CASE. THAT IS, WHAT KIND OF SCIENTIFIC EVIDENCE DID
THEY HAVE WITH THEIR WITNESSES. ALL THE FACTS THAT GO TO THE
STRENGTH OF THE EVIDENCE, THOSE ARE ALL INCLUDED IN THE ANALYSIS
THAT PRODUCED THE FIGURES IN DB-9%5, AND, AND WITH RESPECT TO THE
WAY THE REGRESSION ANALYSIS WORKS AND IN THAT KIND OF CONTEXT.
INDEED THESE DISPARITIES ARE DISPARITIES ACROSS ALL THE CASES IN
CATEGORIES WHICH CONCEPTUALLY ARE EQUIVALENT TO BEING MATCHED ON
ALL THESE BACKGROUND FACTORS. INCLUDING ALL THESE EVIDENTIARY
FACTORS.
THE COURT: ALL RIGHT. WELL, I HAVE YOUR TESTIMONY.
WHAT ARE WE ON, 267
MR. BOGER: THATS RIGHT. YOUR HONOR.
THE COURT: ALL RIGHT. I WOULD NOTE FOR THE RECORD
JUST AS A NOTE TO MYSELF, AND TO YOU ALL, I THINK I SAID ROMAN
NUMERAL II LINE, AND 3D LINE ARE PROBABLY THOSE I WOULD BE, WHAT
I WOULD BE MOST INTERESTED IN AND BECAUSE OF MANY CONCERNS,
INCLUDING THE ONE I WAS JUST DISCUSSING WITH PROFESSOR BALDUS, 1
AM TROUBLED BY THIS EXHIBIT, BUT I THINK IT IS SOPHISTICATED
ENOUGH TO ALLOW IT IN AND THEN WE LL JUST HAVE TO LOOK AT THE
WEIGHT OF IT IN THE CONTEXT OF WHAT FOLLOWS.
IT’S A GOOD TIME FOR A LUNCH BREAK.
Pd
8
a
B
d
e
,
BE
CO
927
BALDUS - DIRECT
DO YOU OUT-OF-CITY LAWYERS HAVE TRAVEL PLANS WHICH
WOULD, ARE YOU TRYING TO CATCH A PLANE TO GO HOME, AND WOULD YOU
LIKE TO LEAVE AT 4:307 I DON’T WANT TO DO IT JUST TO KNOCK OFF
EARLY BUT IF YOU’RE TRYING TO CATCH A PLANE AND TO KNOCK OFF AT
4:30 WOULD HELP, I WOULD CONSIDER IT.
MR. BOGER: I APPRECIATE IT. YOUR HONOR. WE HAD
ANTICIPATED WE WOULD BE INTO NEXT WEEK AND WE ARE HERE FOR THE
WEEKEND.
THE COURT: ALL RIGHT. WELL PLAN ON RUNNING ON UP TO
F100 OR 5:30 TONIGHT.
| WELL BE IN RECESS UNTIL QUARTER OF TWO.
(COURT RECESSED FOR LUNCH.)
THE COURT: DO YOU HAVE A PREDICTION OF WHEN YOULL BE
THROUGH WITH YOUR DIRECT CASE?
MR. BOGER: DIRECT CASE OR PROFESSOR BALDUS, YOUR
HONOR?
THE COURT: DIRECT CASE.
MR. BOGER: I THINK THAT PROFESSOR BALDUS- DIRECT THIS
AFTERNOON WILL TAKE ANOTHER TWO, TWO AND A HALF HOURS, AT THE
QUTSIDE.
AND THEN I SUSPECT PROFESSOR WOODWORTH WILL BE HALF A
DAY, TO THREE-QUARTERS QF A DAY.
UNDER YOUR HONOR‘S SUGGESTION THAT PROFESSOR BERK BE A
REBUTTAL WITNESS, WHICH WE HAVE CONCURRED IN, WE MIGHT HAVE ONE
BALDUS - DIRECT
-
OTHER SHORT WITNESS IN ADDITION TO THE MCCLESKEY TESTIMONY
< RELATING TO, TO THE INEFFECTIVE ASSISTANCE OF COUNSEL. S50 I
3 | WOULD SUSPECT THAT WE WOULD BE GOING INTO TUESDAY. MAYBE EVEN
4 INTO THE BEGINNING OF WEDNESDAY, IF THERES A SUBSTANTIAL
| 3S CROSS-EXAMINATION OF PROFESSOR BALDUS OR PROFESSOR WOODWORTH.
® & [THE COURT: I THINK I WOULD LIKE TO COMPARTMENTALIZE
7 |THE TRIAL SO WE DON’T DO THE INEFFECTIVE ASSISTANCE OF COUNSEL
8 ISSUE UNTIL WE“RE COMPLETELY -—
? MR. BOGER: IT WILL BE INTO TUESDAY. 4:00, 4:30. I
10 DON’T KNOW WHETHER MS. WESTMORELAND COULD FINISH HER CROSS TODAY
11 EVEN IF WE GET WELL UNDERWAY. I REALLY HAVE NO IDEA. [I SUSPECT
12 WE“D BE INTO MONDAY ON THE CROSS-EXAMINATION OF PROFESSOR
13 BALDUS.
14 I THINK DIRECT AND CROSS OF PROFESSOR WOODWORTH IS
15 PROBABLY A DAY. AGAIN DEPENDING ON A ROUGH ESTIMATION OF CROSS.
16 : THE COURT: OKAY. HOW LONG DO YOU THINK YQU-RE GOING
17 TO BE?
18 MS. WESTMORELAND: ON OUR CASE. YDUR HONOR, OR ON —-—
w 19 THE COURT: YOU WANT TQ ESTIMATE AS TO HOW LONG YOU ARE
20 GOING TO CROSS PROFESSOR BALDUS?
21 MS. WESTMORELAND: DEFINITELY I WOULD NOT FINISH WITHIN
22 AN HOUR THIS AFTRERNOON. THAT'S AR FAIR ESTIMATE, YOUR HONOR.
23 AND PROBABLY. DEPENDING ON HOW FAR WE DO GET TODAY, I DONT
24 ANTICIPATE THAT IT WOULD TAKE THE WHOLE DAY MONDAY. OTHER THAN
23 THAT, I’M A LITTLE AT A LOSS AT THIS POINT TO DEFINE HOW LONG IT
146
P27
BALDUS - DIRECT
| WOULD TAKE. PERHAPS THREE-FOURTHS OF A DAY IS A ROUGH ESTIMATE.
FOR THE PRESENTATION OF THE CASE, FOR THE PRESENTATION
OF THE CASE.» I WOULD ANTICIPATE. NOT COUNTING THE
|
CROSS~-EXAMINATION, PROBABLY TWO TO THREE DAYS.
| THE COURT: ALL RIGHT. I’M GOING TO TAKE NEXT
WEDNESDAY AFTERNOON OFF. I“VE GOT TO PRETRY A CASE THEN SO IT
WILL BE READY FOR TRIAL IN SEPTEMBER. WE WON‘T RUN WEDNESDAY
| AFTERNOON.
ALL RIGHT. GO AHEAD.
DAVID C. BALDUS,
BEING PREVIOUSLY DULY SWORN, RESUMED THE WITNESS STAND AND
TESTIFIED FURTHER AS FOLLOWS:
DIRECT EXAMINATION (CONT-D)
BY MR. BOGER:
@. PROFESSOR BALDUS, --
MR. BOGER: =-- ACTUALLY THIS I3 DIRECTED TO THE COURT,
IM AFRAID I MAY HAVE CONFUSED THE RECORD MORE THAN 1 CLARIFIED
IT BY OFFERING A SUBSTITUTE DOCUMENT WHICH IS HARD TQ GET INTO
THE DOCUMENT FOLDERS.
YOUR HONOR HAD SAID THAT A FORTION OF THE, DB-96 WHICH
WAS MOST PROBATIVE AND OF MOST INTEREST WAS THE PORTION ROMAN
NUMERAL II WHICH I TAKE IT, YOUR HONOR, TO BE THE FORTION THAT
DEALT WITH STATISTICALLY SIGNIFICANT "P" IS LESS THAN .10,
NON-ARBITRARY FACTORS, THAT COLUMN. UNFORTUNATELY. ON OUR
a
)
BALDUS - DIRECT
IY
(REVISED VERSION THAT GOT MISLETTERED AS ROMAN NUMERAL I-~C. SO
2 THERE REALLY IS NO ROMAN NUMBERAL II ON THE SECOND VERSION.
3 | THE COURT: I THINK I SUCCEEDED IN PUTTING THE BAD ONES
4 BACK IN THE BOOK IN BOTH CASES.
2 MR. BOGER: WELL, THE BAD ONE MAY HAVE BEEN THE GOOD
ae & ONE, UPON SECOND THOUGHT. AS LONG AS THE RECORD IS CLEAR THAT
7 IN THE REVISED DB-96 IT’S I-C, I BELIEVE. AND IN THE EARLIER
8 VERSION WAS ROMAN NUMERAL II, --
4 THE COURT: OKAY. WELL. THATS, I THINK WE‘RE TALKING
10 ABOUT THE SAME THING.
11 | MR. BOGER: ALL RIGHT.
12 BY MR. BOGER?
13 @. PROFESSOR BALDUS, LET ME ASK YOU ONE QUESTION ABOUT THE I-C,
14 REVISED VERSION, ROMAN NUMERAL II IN THE OTHER VERSION.
15 THE JUDGE HAD ASKED YOU PRIOR TO THE LUNCH BREAK
16 WHETHER RACE, EXCUSE ME, WHETHER STRENGTH OF EVIDENCE FACTORS
17 WERE INCLUDED WITHIN THAT COLUMN.
| 18 DO YOU HAVE AN ANSWER FOR THAT, TO THAT QUESTION NOW?
Ww i . ih. YES.
20 THE COURT: WE WERENT TALKING ABOUT THESE EXHIBITS
2 WITH THAT CONCERN, WERE WE?
22 MR. BOGER: MY IMPRESSION IS, YOUR HONOR. THAT YOU WERE
23 INTERESTED IN WHETHER STRENGTH OF EVIDENCE WAS INVOLVED THERE.
24 I MAY HAVE MISUNDERSTOOD.
23 THE COURT: YES. ALL RIGHT. YES. OKAY.
BALDUS ~ DIRECT
1 | THE WITNESS: YES, YOUR HONOR ASKED ME ORIGINALLY ABOUT
[ THE ANALYSIS REFLECTED IN 3D, AND I INDICATED WE HAD NO
3 INDEPENDENT MEASURES ON THE STRENGTH OF THE EVIDENCE IN THE
4 PROCEDURAL REFORM STUDY.
HOWEVER, IN THE CHARGING AND SENTENCING STUDY WE HAVE A (4
6 SUBSTANTIAL VARIETY OF SUCH MEASURES, AND THE GROUP OF VARIABLES
7 THAT WERE USED AS BACKGROUND CONTROLS AT WHAT IS NOW I-C AND
a WHAT WAS FORMERLY ROMAN NUMERAL 11 ON DB-94, INCLUDED THREE SUCH
7 MEASURES.
10 THE ONE WAS A COUNT, A VARIABLE WHICH COUNTED THE
11 NUMBER OF MAJOR INCRIMINATING EVIDENTIARY FACTORS IN THE CASE.
12 AND ANOTHER ONE WAS WHETHER OR NOT THERE WAS AN I.D. WITNESS OF
13 THE CO-PERPETRATOR AT OR NEAR THE SCENE OF THE CRIME. AND THEN
14 ANOTHER ONE WHICH DID NOT MEASURE STRENGTH OF THE EVIDENCE
15 DIRECTLY, BUT IS A PROXY FOR IT, AND THAT IS WHETHER OR NOT THE
16 CO-PERPETRATOR RECEIVED A LESSER SENTENCE. THAT VARIABLE
17 FIGURES IN A SERIES OF OUR ANALYSES AS AN AGGRAVATING FACTOR.
13 AND WHAT I BELIEVE IT REPRESENTS IS THAT IN MOST OR A BIG
w» 19 PROPORTION OF THOSE CASES, THE CO-PERPETRATOR PROVIDED EVIDENCE
=0 AGAINST THE DEFENDANT.
<1 THAT IS NOT A DIRECT MEASURE, THAT IS ONLY A FROXY FOR
22 IT. BUT THE OTHER TWO ARE DIRECT MEASURES TAKEN SRAUARELY FROM
23 THE FILES OF THE PAROLE BOARD WITH RESPECT TO WHAT THE IMPORTANT
<4 EVIDENTIARY FACTORS WERE.
25 SO THAT INDEED THE SUGGESTION THAT AS FOR THIS
BALDUS ~ DIRECT {
|
|
[DECISION THOSE EVIDENTIARY
FACTORS ARE IMPORTANT 13 RECENTLY
| BORNE OUT BY THE ANALYSIS THAT WE DID IN THIS PARTICULAR
STAGE
jo THE REPORT.
THE COURT: ALL RIGHT. LET/S GO AHEAD.
| MR. BOGER: FINE.
BY MR. ROGER?
12 PROFESSOR paLDUS, DID you CONDUCT ANY ANALYSIS OF THE
DECISION BY THE JURY USING LOGISTIC ANALYSIS TO DETERMINE
WHETHER THERE WERE RACE OF THE DEFENDANT DISPARITIES
AS WELL?
A. YES. WE DID. WE CONDUCTED AN ANALYSIS OF THE JURY DECISION
e
e
e
o
—
—
—
—
—
—
—
—
—
—
i
n
s
MAKING WITH LOGISTIC REGRESSION
PROCEDURES.
(Ge LET ME DIRECT YOUR ATTENTION TO DB-97 MARKED FOR
{
| IDENT IFICATION. can YOU IDENTIFY THAT DOCUMENT?
a. ves. UE-97 1% 6 TAGE 48 FROM OUR REPORT, AND IT REFLECTS
he RESULTS OF SEVEN SUCH ANALYSES.
o can YOU BRIEFLY TELL US WHAT THE RESULTS. OR ACTUALLY WHAT
(ANALYSES YOu CONDUCTED, AND THEN WHAT THE RESULTS ARE?
a. VES. PART 1 REFLECTS THE ANALYSIS FROM THE CHARGING AND
| SENTENCING STUDY. AND PART 2 REFLECTS THE RESULTS FROM THE
| PROCEDURAL REFORM STUDY.
AND IT INDICATES
THAT IN THE CHARGING AND SENTENCING
STUDY THAT THERE ARE RACE OF ViCTIM EFFECTS cSTIMATED
IN EACH
|
:
OF THESE ANALYSES.
HOWEVER: IN ONLY ONE OF THEM po THOSE
Frrecys RISE TO A LEVEL OF STATISTICAL
SIGNIFICANC
E BEYOND THE
|.03 LEVEL.
|
|
—
—
—
—
—
c
m
s
CE
e
m
—
—
—
—
—
—
—
E
S
P
—
BALDUS - DIRECT
Pa
y IN OUR ANALYZES OF JURY DECISION MAKING, WE FOUND
FREGUENTLY THAT WHEN WE ONLY INCLUDED A HANDFUL OF VARIABLES,
RACE EFFECT IS NOT APPARENT: AND IT GENERALLY DOES NOT APPEAR
UNTIL THERES A SIGNIFICANTLY LARGER GROUP OF BACKGROUND
CONTROL VARIABLES.
THE RACE OF VICTIM EFFECTS PRESENTED IN PART 2 OF THAT
TABLE ARE DRAWN FROM THE PROCEDURAL REFORM STUDY, AND THOSE
ANALYSES SHOW RACE OF VICTIM EFFECTS. BUT THEY ARE NOT AS STRONG
I
P
o
l
RE
T
E
«S
RR
i
Be
AS THE RACE OF VICTIM EFFECTS THAT WERE SHOWN IN THE LEAST
po
h o SRUARES REGRESSION ANALYSES.
11 WHEN ONE TURNS TO COLUMNS D AND E OF DB-97, HERE AGAIN
12 YOU CAN SEE A PATTERN IN WHICH THE CO-EFFICIENTS ARE UNSTABLE.
13 AND THEY TEND, IN FACT, OVERALL TO SUPPORT THE PROPOSITION THAT
14 CONTROLLING FOR ALL THE BACKGROUND FACTORS AND THE RACE OF THE
15 VICTIM, THAT BLACK OFFENDERS ARE AT AN ADVANTAGE WHEN ONE VIEWS
16 THE STATE AS A WHOLE AT THE PENALTY TRIAL PHASE, ALTHOUGH NONE
37 OF THESE CO-EFFICIENTS REACHES A SUBSTANTIAL LEVEL OF
13 STATISTICAL SIGNIFICANCE.
A 12? A. LET ME ASK YOU, PROFESSOR BALDUS, ONCE AGAIN, DOES, DO THESE
20 CO-EFFICIENTS REFLECT THE RESULTS OF ANY DISCRIMINATION THAT
21 MIGHT HAVE OCCURRED OR ANY DISPARITIES THAT MIGHT HAVE RESULTED
22 IN EARLIER PHASES IN THE SYSTEM?
23 A. NO. THEY DO NOT TAKE INTO ACCOUNT ANY EFFECTS OTHER THAN
<4 THOSE PRODUCED AT THE PENALTY TRIAL ITSELF. THEY INDICATE THAT
23 THE. WITH RESPECT TO THE RACE OF THE DEFENDANT. THAT PLAINLY THE
— — — — —— | — ——— ——— ——" ——p— — —
934
BALDUS - DIRECT
OVERALL EFFECTS THAT WE-VE SEEN IN CUR COMPARISONS OF DEATH
SENTENCING RATES OF OFFENDERS INDICTED FOR MURDER ARE NOT THE
PRODUCT OF JURY SENTENCING DECISIONS WITH RESPECT TO THE RACE OF
THE DEFENDANT.
THEY ALS0 SUGGEST THAT WITH RESPECT TO THE RACE OF THE
VICTIM THAT THE EFFECT THERE IS ONLY PARTIALLY THE PRODUCT OF A
JURY DEATH SENTENCING DECISION AT PENALTY TRIALS.
@. NOW WITH RESPECT TO THE DIFFERENT ANALYSES YOU RAN, USING
THIS LOGISTIC METHOD. ARE THE FIVE LEGITIMATE FACTORS IN 1A THE
SAME THAT YOU TESTIFIED TO PREVIOUSLY?
A. YES, THEY ARE.
2. AND THE NINE LEGITIMATE FACTORS, YOU/VE PREVIOUSLY GIVEN
INDICATION ABOUT NINE AT ONE POINT. IS IT THE SAME GROUP THAT'S
EMPLOYED HERE?
A. YES. THOSE ARE RESPECTIVELY THE FIVE LEGITIMATE FACTORS
THAT ARE IN TABLE 22 OF THE, OUR REPORT, AND THE NINE FACTORS
ARE IN TABLE 222A OF THAT REPCRT. u
@. NOW, IN IC YOU MENTIONED 146 STATISTICALLY SIGNIFICANT
NON-ARBITRARY FACTORS.
MY RECOLLECTION IS NOT CLEAR ON WHETHER YOUVE ACTUALLY
REFERRED TO THIS GROUP PRIOR TO NOW IN YOUR TESTIMONY.
A. NO, I HAVE NOT. THOSE ARE IN SCHEDULE ¥?., OF THE TECHNICAL
APFENDIX OF OUR REPORT.
@. THAT TECHNICAL APPENDIX HAS BEEN MARKED AS DB-956A FOR
IDENTIFICATION. COULD YOU DIRECT US TO WHAT THOSE 16 ARE. USING
kJ
8
«£
3 LJ
rR
BALDUS -~ DIRECT
THAT APPENDIX?
A. YES. WOULD YOU LIKE ME TO READ WHAT THOSE FACTORS ARE.
COUNSEL?
lo. IF YOU WOULD, QUICKLY?
A. YES. THE FIRST ONE -- FIRST, I“M READING FROM SCHEDULE ¢ OF
THE TECHNICAL APPENDIX.
THE COURT: I DON’T HAVE THAT.
MR. BOGER: I THOUGHT WE PASSED A COPY UP TU YOUR
HONOR. I THINK THE CLERK HAS A COPY.
THE COURT: I HAVE 8 AND I HAVE 10 -- OKAY, IT’S JUST
OUT OF ORDER.
MR. BOGER: SORRY FOR THAT, YOUR HONOR.
THE WITNESS: MAY I PROCEED?
BY MR. BOGER:
@. OF COURSE.
A. THE FACTORS HERE THAT ARE IMPORTANT IN PREDICTING JURY DEATH
SENTENCING DECISIONS THAT WERE PRODUCED WITH THIS ANALYSIS ARE
THE FOLLOWING.
WHETHER OR NOT THE DEFENDANT WAS THE TRIGGER MAN,
WHETHER OR NOT A B2 STATUTORY ACGRAVATING
CONTEMPORANEOUS OFFENSE WAS PRESENT.
WHETHER THE DEFENDANT HAD A Bl PRIOR RECORD.
WHETHER OR NOT THE VICTIM WAS WITH OR WITHOUT CLOTHING.
MULTIPLE HEAD SHOTS.
WHETHER THE DEFENDANT HAD BEEN RELEASED FROM PRISON {
ha
+
O
36
BALDUS - DIRECT
WITHIN THE PRECEDING YEAR.
WHETHER DEFENDANT CLAIMED COERCION.
| WHETHER THERE WAS A REVENGE MOTIVE.
WHETHER THE DEFENDANT HAD A HISTORY OF DRUG ABUSE.
WHETHER THE DEFENDANT HAD A MITIGATING MOTIVE.
WHETHER THERE WAS AN INSURANCE MOTIVE.
THE NUMBER OF MAJOR AGGRAVATING. NON-STATUTORY
| AGGRAVATING FACTORS IN THE CASE.
DEFENDANT “3S AGE AT THE TIME OF FIRST FELONY ARREST.
DEFENDANT INVOKED THE SANITY DEFENSE OR NOT.
DEFENDANT PANICKED IN THE COURSE OF A BURGLARY.
AND THE NUMBER OF MAJOR AND INCRIMINATING EVIDENTIARY
FACTORS IN THE CASE.
@. HOW WERE THESE 14 FACTORS DERIVED?
‘A. THESE WERE DERIVED IN THE SAME MANNER THAT I REFERRED TO
EARLIER WITH RESPECT TO THE ABBREVIATED LIST THAT WAS USED IN
THE ANALYSIS OF THE PROSECUTORIAL DECISIONS.
WE IDENTIFIED FACTORS THAT WERE STATISTICALLY
SIGNIFICANT AT THE .10 LEVEL IN EXPLAINING THE JURY SENTENCING
DECISIONS.
WE FURTHER LIMITER THE GROUP OF VARIABLES TO THOSE THAT
WERE LEGITIMATE AGGRAVATING AND MITIGATING FACTORS, AND FURTHER
TO AGGRAVATING FACTORS THAT HAD AN AGGRAVATING EFFECT IN THE
ANALYSIS, AND TO MITIGATING FACTORS THAT HAD A MITIGATING EFFECT
IN THE ANALYSIS.
O
E
W
M
A
e
g
Pr
y
—
-
id
13
BALDUS - DIRECT
'Q. FINALLY. LET ME ASK YOU ABOUT PART 2; THE FIVE AND NINE
LEGITIMATE VARIABLES, RESPECTIVELY. IN "A" AND "B", ARE THOSE
THE FIVE AND NINE TO WHICH WE PREVIOUSLY REFERRED?
A. I AM NOT SURE THAT THEY ARE THE. THE FIVE VARIABLES REFERRED
THERETO ARE ON SCHEDULE 10 OF THE TECHNICAL APPENDIX: THE
NINE-VARIABLE LIST IS FOUND IN SCHEDULE 11 OF THE TECHNICAL
APPENDIX: AND THE ELEVEN STATUTORIALLY SIGNIFICANT NON-ARBITRARY
FACTORS THAT I REFERRED TO IN THE PART 2 HERE, WHICH RELATES TO
THE PROCEDURAL REFORM STUDY, ARE FOUND IN SCHEDULE 3 OF THE
TECHNICAL APPENDIX.
MR. BOGER: YOUR HONOR, FOR THE CLARITY OF THE RECORD
WE CAN EITHER HAVE PROFESSOR BALDUS GO THROUGH THOSE OR WE CAN
ADMIT GENERALLY THE TECHNICAL APPENDIX, WHICH ITSELF. I THINK
ADDS NO MORE EVIDENCE. BUT SIMPLY REFLECTS WHAT HE“S RELIED UPON
IN 97 AND EARLIER IN 96.
THE COURT: ALL RIGHT, LET”S DO THAT. DO YOU HAVE ANY
OBJECTION, MS. WESTMORELAND?
MS. WESTMORELAND: NOT FOR THAT PURPOSE. YOUR HONOR.
THE COURT: ALL RIGHT.
MR. BOGER: YOUR HONOR, I MOVE THE ADMISSION OF DB-24A
INTO EVIDENCE AND MOVE AT THE SAME TIME THE ADMISSION OF DB-97
INTO EVIDENCE.
MS. WESTMORELAND: YOUR HONOR, IN ADDITION TO THE SAME
OBJECTIONS: I WOULD NOTE THAT I THINK THE SAME CONCERNS ARE
PRESENT IN THIS TABLE, AS IN THE PREVIOUS TABLE, WHETHER THESE
a3
BALDUS - DIRECT
1 PARTICULAR FACTORS THAT ARE BEING ACCOUNTED FOR ACTUALLY HAVE
< ANY RELEVANCE TO THE ISSUE THAT IS BEING PRESENTED IN THE CASE.
3 FOR THE MERE CONSIDERATION OF FIVE FACTORS OR NINE FACTORS OR
SIXTEEN FACTORS 1S SUFFICIENT TO SHOW, MAKE A STATISTICAL
SHOWING NECESSARY. :
MR. BOGER: YOUR HONOR, I THINK PROFESSOR BALDUS’
TESTIMONY HAS BEEN HES EMPLOYED A NUMBER OF DIFFERENT METHODS,
AS YOU KNOW, TO TRY TO SEE WHETHER THE RACIAL FACTORS CAN BE
3
©
~
Oo
4
0
»
ACCOUNTED FOR AND THE EXTENT TO WHICH THEY CAN BE, HES NOT
10 RELIED ON ANY ONE METHOD AND INDEED HIS WHOLE APPROACH HAS BEEN
11 0 SET AS MANY METHODS AS HE COULD BEFORE THE COURT. AND THESE
i2 ARE SIMPLY OTHER METHODS THAT HE“S USED AND ON WHICH HE’S RELIED
13 IN REACHING HIS FINAL CONCLUSIONS. I DO THINK THEY'RE QUITE
14 RELEVANT, ALTHOUGH PERHAPS NONE OF THEM IN ITSELF IS SUFFICIENT.
13 | THE COURT: I WILL ALLOW IT, NOTING THAT I, AT THIS
146 MOMENT. HAVE LESS CONFIDENCE IN THIS THAN I DO IN SOME OF THE
17 OTHERS I“VE SEEN. NO USE GOING INTO THAT.
13 S0 97 WILL BE ADMITTED.
w 19 MR. BOGER: THANK YOU, YOUR HONOR.
20 BY MR. BOGER:
<1 a. PROFESSOR BALDUS,. DID YOU CONDUCT ANY ANALYSES THAT LOOKED
22 AT THE COMBINED EFFECTS OF BOTH THE PROSECUTORIAL DECISION TO
23 SEEK A DEATH SENTENCE AND THE JURY DECISION ON WHETHER TO IMPOSE
24 A LIFE OR DEATH SENTENCE?
(4
)
ra
Wc
J
ai
r
0
TE
S
B
R
i
10
939
BALDUS —- DIRECT
a. LET ME DIRECT YOUR ATTENTION TO WHAT S BEEN MARKED AS DB-%3
FOR IDENTIFICATION. CAN YOU IDENTIFY THAT DOCUMENT?
A. YES. DB-98 IS TABLE 48A FROM OUR REPORT. AND IT PRESENTS
|THE, THE RESULTS OF A SERIES OF LEAST SQUARES REGRESSION
ANALYSES AND LOGISTIC REGRESSION ANALYSES. AND THIS ANALYSIS
EMPLOYS DATA FROM THE PROCEDURAL REFORM STUDY. AND IT FOCUSES
ON THE COMBINED EFFECTS OF THE DECISION TO ADVANCE THE CASE TO A
PENALTY TRIAL, AND THE JURY SENTENCING DECISION.
SPECIFICALLY, rr ADDRESSES THE RUESTION, AMONG THE
POPULATION OF OFFENDERS CONVICTED OF MURDER AT TRIAL, IS THERE A
DIFFERENCE IN THE RATES AT WHICH OFFENDERS WERE WHITE OR BLACK
VICTIMS ARE SENTENCED TO DEATH, AFTER ADJUSTMENT FOR BACKGROUND
FACTORS.
AND COMMENCING WITH ROW 1. WE CAN SEE THE EFFECTS OF
CONTROLLING FOR A SMALL NUMBER OF CONCEPTUALLY IMPORTANT
BACKGROUND FACTORS. WE SEE THAT IN COLUMN Bs, THE. THERE IS A
NINE PERCENTAGE POINT DIFFERENCE IN THOSE RATES AT WHICH THE TWO
GROUPS RECEIVE DEATH SENTENCE, WITH WHITE VICTIM CASES RECEIVING
AT MORE ELEVATED DEATH SENTENCING RATE.
THE NEXT COLUMN, LABELED "DEATH ODDS MULTIPLIER"
REFLECTS THE NUMBER OF TIMES ONE-S ODDS ARE INCREASED. ARE
ESTIMATED TO INCREASE IN THIS ANALYSIS IF ONE’S VICTIM IS WHITE.
|
|
AND THEN THE FINAL COLUMN PRESENTS THE LOGISTIC
REGRESSION CO-EFFICIENT FROM WHICH THAT WAS CALCULATED.
THE SUCCEEDING ROWS HERE LIST AN EVER INCREASING NUMBER
-~
~~
BB
p
W
a
10
$40
BALDUS — DIRECT
OF STATUTDRY AGGRAVATING, MITIGATING, NON-STATUTORY AGGRAVATING
FACTORS, IN EVER INCREASING NUMBER, AND ONE CAN SEE THAT AS ONE
Fre DOWN THE TABLE, THAT THE OVERALL DISPARITIES PERSIST
REGARDLESS OF THE BACKGROUND FACTORS THAT ARE CONTROLLED FOR
SIMULTANEOUSLY IN THE STATISTICAL ANALYSIS.
AGAIN, THESE ANALYSES PRODUCED FROM THE PROCEDURAL
REFORM STUDY SHOW THAT WHEN YOU COMBINE THE IMPACT OF THE TWO
FACTORS, TWO DECISION POINTS, THAT THERE IS A STRONG AND
PERSISTENT RACE OF VICTIM EFFECT.
Gi. AND THAT DOES NOT TAKE ANY ACCOUNT OR DOES IT, PROFESSOR
BALDUS, OF ANY EFFECTS, THE DISPARITIES THAT MAY HAVE OCCURRED
EARLIER IN THE SYSTEM?
A. NO, IT DOES NOT. THESE ARE JUST THE EFFECTS ASSOCIATED WITH
THE TWO DECISION POINTS THAT I JUST REFERRED TO, THE DECISION TO
ADVANCE THE CASE TO PENALTY TRIAL AND THE DEATH SENTENCING
DECISION.
MR. BOGER: YOUR HONOR. I MOVE THE ADMISSION OF DB-98
INTO EVIDENCE.
MS. WESTMORELAND: SAME OBJECTIONS AS PREVIOUSLY NOTED.
YOUR HONOR. |
THE COURT: ALL RIGHT, ILL RECEIVE IT.
MR. BOGER: THANK YOU, YQUR HONOR.
BY MR. BOGER:
QR. PROFESSOR BALDUS, YOU’VE INDICATED THROUGHOUT YOUR TESTIMONY
NOW, THAT RACIAL EFFECTS, RACE OF THE VICTIM, RACE OF THE
241
BALDUS — DIRECT
1 |DEFENDANT, HAVE AN ACTUAL IMPACT ON GEORGIA‘S CHARGING AND
2 SENTENCING SYSTEM.
3 ARE THERE ANY MEANS THAT YDU KNOW OF WHEREBY TO
4 ESTIMATE THE LIKELY SENTENCING OUTCOMES IF RACIAL FACTORS DID
= NOT PLAY ANY ROLE IN THE GEORGIA SYSTEM?
1) A. ONE CAN NEVER KNOW FOR SURE, OF COURSE. WE MADE AN ATTEMPT
7 TO ESTIMATE THE RESULTS THAT ONE WOULD LIKELY SEE IF THE
a EXERCISE OF DISCRETION IN THE SYSTEM WERE REDUCED.
? @. HOW DID YOU GO ABOUT THAT ESTIMATION?
Q A. WE DID IT IN THIS WAY. WE HYPOTHESIZED THE SYSTEM IN WHICH
11 ALL DEFENDANTS CONVICTED OF MURDER WOULD HAVE THEIR CASES
12 AUTOMATICALLY ADVANCED TO A PENALTY TRIAL AND THE JURY
13 SENTENCING THEM ON A LIFE OR DEATH DECISION, WOULD SENTENCE THEM
14 AT THE AVERAGE RATE WHICH JURIES CURRENTLY SENTENCE OFFENDERS
15 WHILE CONTROLLING FOR THE VARYING AGGRAVATION LEVELS OF THE
15 CASES THAT WERE ADVANCED TO THIS HYPOTHETICAL PENALTY TRIAL.
37 NOW THIS WAS DONE SPECIFICALLY BY TAKING THE CASES IN
13 WHICH A MURDER CONVICTION HAD BEEN OBTAINED AND DETERMINING HOW
» 1? LIKELY IT WOULD BE THAT A JURY. FACED WITH THE FACTS OF THAT
20 CASE, WOULD IMPOSE A DEATH SENTENCE IN THAT CASE.
21 SPECIFICALLY, =-—
22 R. EXCUSE ME. HOW DID YOU MAKE THAT -—
232 A. WE MADE THAT DETERMINATION BY EXAMINING THE RESULTS OF THE
24 REGRESSION ANALYSIS WHICH ARE INCLUDED IN SCHEDULE 2? OF THE
2% TECHNICAL APPENDIX.
——— ——— —— ——_ —— ————— —— — —— —— » : Se
942
BALDUS - DIRECT
THOSE WERE THE FACTORS THAT BEST EXPLAINED THE JURY'S
SENTENCING DECISION, THE LEGITIMATE FACTORS BEST EXPLAINING THE
| JURY DEATH SENTENCING DECISION.
AND ON THE BASIS OF THOSE FACTORS, AND THE
CO-EFFICIENTS THAT WERE ESTIMATED FOR THOSE VARIABLES. WE WERE
ABLE TO EXAMINE EACH OF THESE OFFENDERS IN THE SYSTEM THAT WERE
SENTENCED TO,» CORRECTION. WHO WERE CONVICTED OF MURDER. EITHER
BY PLEA OR AT TRIAL, AND DETERMINE HOW SERIOUS THOSE CASES WOULD
BE CONSIDERED IN THE EYES OF A JURY, THE EYES OF THE AVERAGE
JURY IN GEDRGIA, IF YOU WILL, AS REFLECTED BY THIS REGRESSION
ANALYSIS WHICH IS SUGGESTED IN SCHEDULE 9, SUGGESTED BECAUSE I
ONLY HAVE LISTED THERE THE VARIABLES THAT WERE INCLUDED IN THE
MODEL.
S50 THIS HAD THE EFFECT OF PRODUCING A BREAKDOWN OF ALL
OF THE OFFENDERS CONVICTED OF MURDER INTO VARIOUS SUB-GROUPS
WHICH INCREASED IN THEIR LEVELS OF AGGRAVATION AS MEASURED BY
THIS ANALYSIS.
MOREOVER. WE KNEW WHAT THE ACTUAL JURY DEATH SENTENCING
RATE WAS AMONG EACH OF THOSE SUB-GROUFS OF CASES BY VIRTUE OF
WHAT THE JURIES HAD ACTUALLY DONE IN THE 253 CASES WHERE THEY
HAD PASSED OM, PASSED ON A LIFE OR DEATH DECISION.
WE WERE THEN ABLE TO ESTIMATE WHAT THAT DEATH
SENTENCING RATE WAS AMONG THOSE ACTUAL CASES AND THEN WE APPLIED
THAT DEATH SENTENCING RATE TO THE HYPOTHETICAL POPULATION OF
CASES THAT WOULD HAVE COME BEFORE A DEATH SENTENCING JURY IN THE
BALDUS - DIRECT
ABSENCE OF INTERVENTION BY THE PROSECUTOR WHICH WOULD HAVE
REDUCED THE NUMBER OF CASES COMING BEFORE THE PENALTY TRIAL
JURIES.
‘Qn. LET ME ASK YOU TO TURN TO DB-99 MARKED FOR IDENTIFICATION.
CAN YOU IDENTIFY THAT DOCUMENT?
A. YES. DB-99 IS TABLE 60 FROM OUR REPORT. AND IT SHOWS THE
RESULTS OF THIS ANALYSIS. THE FIRST ~——
R. WHAT DOES COLUMN B REPRESENT?
A. COLUMN B REPRESENTS THE NUMBER OF DEATH SENTENCES THAT WOULD
OCCUR. WE ESTIMATE WOULD OCCUR UNDER THIS HYPOTHETICAL DEATH
SENTENCING SYSTEM, AND IT ALSO INDICATES THE DISTRIBUTION OF
THOSE DEATH SENTENCES ACCORDING TO THE DEFENDANT-VICTIM RACIAL
COMBINATION.
THE FIRST THING THAT WAS CLEAR FROM THIS HYPOTHETICAL
ANALYSIS WAS THAT IT STRONGLY SUGGESTED THAT IF ALL OF THE CASES
THAT RESULTED IN MURDER CONVICTIONS WERE TAKEN BEFORE A DEATH
SENTENCING JURY, THAT THE NUMBER OF DEATH SENTENCES IMPOSED
WOULD DRAMATICALLY INCREASE. AS ONE CAN SEE BY EXAMINING THE
NUMERATOR OF THE FRACTIONS LISTED IN COLUMN B, THAT ESTIMATED
NUMBER OF DEATH SENTENCES WAS 301.
@. EXCUSE ME, PROFESSOR BALDUS. YOU SAID NUMERATORY
A. PARDON ME. I MEAN DENOMINATOR. PARDON ME. THAT THE
DENOMINATOR WOULD BE 201.
AND WE CAN SEE THAT IN THIS HYPOTHETICAL SYSTEM. THAT
72 OF THOSE 301 DEATH SENTENCES WOULD HAVE BEEN IMPOSED IN CASES
944
BALIWS ~ DIRECT
1 "IN WHICH THE DEFENDANT WAS BLACK AND THE VICTIM WAS WHITE.
2 SIMILARLY, WE OBSERVE IN THIS HYPOTHETICAL SET OF
|
3 RESULTS, 128 DEATH CASES AMONG CASES WHERE —— DEATH SENTENCES.
4 RATHER, AMONG CASES WHERE THE DEFENDANT WAS WHITE, AS WAS THE
VICTIM. (4
AND ONE CAN SEE BY EXAMINING ROWS 1A THROUGH -D THE
DISTRIBUTION OF DEATH SENTENCES BY THE DEFENDANT-VICTIM RACIAL
COMBINATION IN THE HYPOTHETICAL SYSTEM.
9
fo
A
H
,
MOVING DOWN TO PART 2, WE PRESENT THE DISTRIBUTION BY
10 RACE OF VICTIM IN THE HYPOTHETICAL SYSTEM. AND THEN FINALLY IN
11 PART 3, THE DISTRIBUTION IS PRESENTED ACCORDING TO THE RACE OF
12 THE DEFENDANT.
13 NOW, TO BE ABLE TO PRODUCE SOME ESTIMATE OF THE IMPACT
14 THAT THE EXERCISE OF DISCRETION THAT THESE TWO POINTS MIGHT BE
13 HAVING, WE COMPARED THESE HYPOTHETICAL RESULTS WITH THE ACTUAL
14 RESULTS THAT OBTAINED IM THE SYSTEM. AND THOSE ARE PRESENTED IN
17 COLUMN C.
18 AND THERE WE SEE THE RACIAL DISTRIBUTION WITHIN OUR
w 19 FOUR PRINCIPAL CATEGORIES OF THE DEATH SENTENCES ACTUALLY
20 IMPOSED IN OUR SAMPLE.
21 WE THEN COMPARED THOSE TWD, AND THOSE RESULTS ARE
22 PRESENTED IN COLUMN D.
23 AND COLUMN D INDICATES THAT THE NUMBER OF DEATH
24 SENTENCES IMPOSED IN THE CATEGORY OF BLACK DEFENDANT-WHITE
23 VICTIM WOULD DROP TD A POINT THAT THE PROFORTION OF ALL DEATH
BALDUS ~ DIRECT
1 | SENTENCE CASES WITH A BLACK DEFENDANT AND WHITE VICTIM WOULD
DROP FROM 24 PERCENT, PARDON ME, WOULD DROP FROM 3% PERCENT TO
24 PERCENT IN THE HYPOTHETICAL SYSTEM. REPRESENTING A DECREASE
OF FIFTEEN PERCENTAGE POINTS IN THE REPRESENTATION OF THAT CLASS
OF CASE.
Q. IS IT YOUR TESTIMONY, PROFESSOR BALDUS, THAT THE NUMBER OF
~
f
(B
3
Pp
WB
NN
BLACK DEFENDANT-WHITE VICTIM CASES RECEIVING A DEATH SENTENCE
WOULD DROP OR THE PROPORTIONS OR BOTH?
A
U
A. NO, IT WAS THE. IT’S THE PROPORTION THAT WOULD DROP. THE
10 NUMBER. ACTUALLY. WOULD SLIGHTLY INCREASE. AS DO ALL, AS DO
13 DEATH SENTENCES IN EACH CATEGORY OF CASE. IT“S THE PROPORTION
0% THAT WOULD DROP AMONG ALL OF THOSE WHO RECEIVED DEATH SENTENCES.
13 WE ALSO SEE A SLIGHT DROP IN THE PROPORTION OF DEATH
ia SENTENCES THAT WOULD GO TO DEFENDANTS WITH WHITE DEFENDANTS. AND
135 WHITE VICTIMS. BUT AS ROW IC, COLUMN D SUGGESTS, THERE WOULD BE
16 A, AN INCREASE IN THE REPRESENTATION OF BLACK DEFENDANTS WITH
17 BLACK VICTIMS IN THIS HYPOTHETICAL SYSTEM.
ig DROPPING DOWN TO ROW 2, WE CAN SEE THAT WHEN YOU LOOK
w 19 AT IT STRICTLY IN TERMS OF RACE OF THE VICTIM, THAT THERE WOULD
20 EE AN 18 PERCENTAGE POINT DROF IN THE REPRESENTATION OF WHITE
21 VICTIMS IN THIS HYPOTHETICAL DEATH ROW, AND A TWENTY FERCENT --
22 TWENTY PERCENTAGE POINT INCREASE IN THE PROPORTION OF BLACK
<3 VICTIM CASES.
24 FINALLY, THE PROPORTION OF BLACK DEFENDANTS IN THE
23 ENTIRE POPULATION WOULD CHANGE VERY LITTLE, AS IS SUGGESTED BY
p
e
246
BALDUS - DIRECT
ROW 2A AND —-B, ONE PERCENTAGE POINT CHANGE.
Gl. WHAT CONCLUSIONS IF ANY, DO YOU DRAW FROM THIS KIND OF
| TABLE?
A. THIS GIVES ONE A SENSE OF THE PRACTICAL CONSEQUENCES.
PRACTICAL IMPACT THAT THE RACIAL FACTORS APPEAR TO BE HAVING IN
THE SYSTEM IN THE SENSE THAT THE EXERCISE OF THE DISCRETION BY
PROSECUTORS AND EXERCISE OF DISCRETION BY JURIES PRODUCE AN
DVERREPRESENTATION OF WHITE VICTIM CASES ON DEATH ROW. AND IN
PARTICULAR, AN OVERREPRESENTATION OF CASES INVOLVING BLACK
DEFENDANTS AND WHITE VICTIMS. THAT'S WHAT IS SUGGESTED BY THIS
ANALYSIS.
R. HAVE YOU THOUGHT OF ANY OTHER WAY ONE MIGHT HYPOTHESIZE THE
EFFECTS OF THE ABSENCE OF RACIAL FACTORS ON SENTENCING OUTCOMES
IN GEORGIA?
1A. YES.
@. CAN YOU TELL US WHAT THAT ALTERNATIVE HYPOTHETICAL SYSTEM
WOULD DO?
A. YES.
THE COURT: LET. ARE YOU GOING TO ANOTHER TABLE?
MR. BOGER: YES, YOUR HONOR.
THE WITNES3: YES. SIR.
THE COURT: IT IS, IM SURE, TOQ FACILE TO SAY THAT,
THIS, BUT IT MAY BE THE INVERSE OF WHAT YOURE SAYING.
HAVING TAKEN RACE, RACIAL CONSIDERATIONS OUT OF THE
SYSTEM, WE WOULD THEN GIVE THE DEATH PENALTY TO SUBSTANTIALLY
Wr
0
a
R
d
47
BALDUS - DIRECT
MORE PEOPLE, AND BLACK PEOPLE, KILLERS OF KIND OF PERSON, WHICH
IN A WAY SUGGESTS THAT RACE I5 SOMEHOW RESTRAINING THE SYSTEM
FROM IMPOSING THE DEATH PENALTY.
MR. BOGER: THAT-S WHY 1 WAS TRYING TO HURRY ON TO
ANOTHER TABLE AND ANDTHER HYPOTHESIS. BECAUSE THERE ARE
DIFFERENT WAYS IN WHICH ONE CAN LOOK AT WHAT THE RESULTS WOULD
BE. AND I HAD PROPOSED BEFORE WE OFFER EITHER TABLE TO GET TO
THE OTHER ONE, TO SEE WHAT THE OTHER HYPOTHESIS IS, BUT ILL BE
GLAD TO HAVE -—-
THE COURT: 00 AHEAD AND LET’S SEE WHAT THE OTHER ONE
IS.
MR. BOGER: ALL RIGHT.
BY MR. BOGER:
Qt. YOU INDICATED, PROFESSOR BALDUS, THAT YOU HAD ANOTHER
HYPOTHETICAL SYSTEM THAT HAD NO RACIAL FACTORS WITH IMPACT.
CAN YOU TELL US WHAT THAT SYSTEM WOULD DO?
A. YES. WHAT WE HYPOTHESIZED IN THIS SYSTEM, IN DB-100 IS A
SYSTEM IN WHICH THE RATE OF DEATH SENTENCING WOULD NOT INCREASE.
WHAT PRODUCED THAT SUBSTANTIAL INCREASE IN THE NUMBER OF DEATH
SENTENCES THAT WE SAW IN THE PREVIOUS TABLE WAS THE RESTRAINT
IMPOSED UPON THE EXERCISE OF FROSECUTORIAL DISCRETION,
PROSECUTORS SCREEN TWO-THIRDS OF THE MURDER CONVICTIONS QUT OF
THE SYSTEM, AND THAT IS WHY WE HAVE SO MANY DEATH SENTENCES
UNDER THE HYPOTHETICAL SYSTEM.
BUT WE THOUGHT THAT YOU COULD MAKE A SHARPER AND
5
A
y
N
e
10
311
“43
BALDUS - DIRECT
TIGHTER COMPARISON IF I HYPOTHESIZED THE SYSTEM THAT PRODUCED
THE SAME NUMBER OF DEATH SENTENCES AS 1S PRODUCED IN THE CURRENT
SYSTEM. AND THEN ESTIMATE WHAT THE DISTRIBUTION OF THOSE
OFFENDERS WOULD BE BY DEFENDANT-VICTIM RACIAL COMBINATION. AND
THIS WAS DONE VERY SIMPLY BY TAKING THE PRESENT NUMBER OF CASES
AND REALLOCATING THEM ACCORDING TO THE DISTRIBUTION THAT WAS
ESTIMATED IN DB-99 AS A RESULT OF OUR HYPOTHETICAL SYSTEM.
AND WHAT WE SEE IS IN DB-100 THE DISTRIBUTION THAT WE
WOULD EXPECT TO SEE IF THERE WERE 128 DEATH SENTENCES IMPOSED IN
THE HYPOTHETICAL SYSTEM AS CONTRASTED TO THE ACTUAL 128 CASES
THAT WERE IMPOSED. "B" PRESENTS THE HYPOTHETICAL RESULTS. MC
SHOWS THE ACTUAL RESULTS IN TERMS OF THE NUMBERS OF PEOPLE IN
EACH ONE OF THESE CATEGORIES. AND WHAT WE SEE IS. BY
COMPARING "B", COLUMNS B AND C., THE DIFFERENCE WHICH IS
REPRESENTED IN COLUMN I,» WE SEE THAT THERE WOULD BE AN ESTIMATED
NINETEEN FEWER PEOPLE ON DEATH ROW WHO WERE BLACK AND HAD A
WHITE VICTIM, FOUR FEWER PEOPLE WHO WERE WHITE AND HAD A WHITE
VICTIM, BUT AN INCREASE OF TWENTY BLACK OFFENDERS WHO HAD A
BLACK VICTIM, AND AN INCREASE OF 3 PEOPLE WHO HAD. AN INCREASE
OF 3 WHITE DEFENDANTS WHO HAD A BLACK VICTIM. SO THAT WHEN YOU
AGGREGATE THOSE DATA, IT SHOWS THAT THERE WOULD BE A SUBSTANTIAL
DROP IN TERMS OF THE NUMBERS, THAT IS, 23 PEOPLE, IN TERMS OF
THE NUMBERS OF OFFENDERS WHO HAD WHITE VICTIMS.
FINALLY, WHEN ONE LOOKS AT ROW 3, YOU SEE THAT THERE
WOULD BE A CHANGE OF ONLY ONE IN TERMS OF THE NUMBERS OF BLACK
Fa
.
O
s
P
c
42
BALIUS - DIRECT
| DEFENDANTS ON DEATH ROW. BUT THE INTERESTING POINT IS THAT THE
IDENTITY OF THOSE BLACK OFFENDERS ON DEATH ROW WOULD BE
SUBSTANTIALLY CHANGED. BECAUSE THERE WOULD BE NINETEEN BLACK
OFFENDERS WHO WOULD NOT BE THERE. THOSE OFFENDERS WHOSE VICTIMS
WERE WHITE, AND THEY WOULD BE REPLACED RY ANOTHER GROUP OF
OFFENDERS WHO RECEIVED LIFE SENTENCES, SOME TWENTY OF THEM. WHO
WOULD NOW FIND THEMSELVES ON DEATH ROW, ACCORDING TO THIS
ANALYSIS OF OUR HYPOTHETICAL SYSTEM.
Q. ARE YOU ABLE, PROFESSOR BALDUS, TO SAY WHICH OF THE TWO
ALTERNATIVE HYPOTHESES MIGHT OBTAIN IF THERE WERE NO RACIAL
FACTORS AT WORK IN THE SYSTEM?
A. I’M SORRY, COUNSEL, WHICH OF THE TWO? R. WHICH OF THOSE TWO HYPOTHESES WOULD ACTUALLY CONTROL OR
Govern THE SYSTEM. IF THERE WERE NO RACIAL FACTORS? ARE YOU
ABLE TO GIVE SUCH TESTIMONY?
A. WELL, THIS, THIS ANALYSIS INDULGES TWO ASSUMPTIONS. ONE,
THAT THE PROSECUTORS HAVE NO CONTROL OVER WHO GDES TO A PENALTY
TRIAL. NUMBER ONE. AND NUMBER TWO, THAT THE JURIES WILL TREAT
THEM IN AN EVENHANDED FASHION. AND THOSE ARE THE TWO
ASSUMPTIONS THAT UNDERLIE THIS ANALYSIS.
ITS FOSSIBLE TO DO THIS ANALYSIS UNDER A, DIFFERENT
SETS OF ASSUMPTIONS. BUT THAT'S, THOSE ARE THE TWO THAT UNDERLIE
THIS, SO I CAN ONLY SPEAK TO THE HYPOTHETICAL RESULTS PRODUCED
IN THIS WAY.
1 CAN SUGGEST TO YOU THAT WE COULD HAVE DONE AN
230
BALDUS - DIRECT
1 ALTERNATIVE ANALYSIS IN WHICH THE, INSTEAD OF USING THE AVERAGE
2 DEATH SENTENCING RATE. OR USE THE WHITE VICTIM DEATH SENTENCING
3 RATE, USE THE BLACK VICTIM DEATH SENTENCING RATE. IF YOU USE
THE WHITE VICTIM DEATH SENTENCING RATE, THE NUMBERS OF PEOPLE ON
DEATH ROW COES UP, HIGHER THAN WHAT WE HAVE HERE.
WHEN YOU USE THE BLACK VICTIM DEATH SENTENCING RATE, IT
GOES DOWN.
THE POINT, HOWEVER. IS THAT REGARDLESS OF WHICH DEATH
Ww
08
0
o
N
0
R
W
SENTENCING RATE YOU USE, THE DISTRIBUTION OF DEATH ROW PERSONNEL
10 UNDER THE HYPOTHETICAL SYSTEM IS THE SAME AS IS ESTIMATED IN
11 THIS HYPOTHETICAL CONTEXT.
12 '@. IN OTHER WORDS. NO MATTER WHICH OF THE HYPOTHETICALS YQOU
13 USE, THE PERSONNEL. TO USE YOUR TERM, ON DEATH ROW WOULD CHANGE?
14 A. YES, THAT'S RIGHT.
13 MR. BOGER: YOUR HONOR, AT THIS TIME. I MOVE THE
16 ADMISSION OF DB-99 AND DB-100 INTO EVIDENCE.
17 MS. WESTMORELAND: YOUR HONOR, ILL OBJECT TO BOTH OF
13 THESE EXHIBITS. [I FAIL TO SEE THE RELEVANCY WHATSOEVER OF THE
12 HYPOTHETICAL SYSTEM AS PROPOSED.
20 THE COURT: I THINK WE HAVE GONE FROM THE STONE AGE TO
21 THE IRON AGE TO THE MODERN AGE TO THE TWILIGHT SEASON.
22 IM NOT SURE WHETHER YOU WANT THESE IM AND I“M NOT SURE
23 WHETHER I WANT THEM IN, AND IM NOT SURE WHAT THEY PROVE.
24 YOU KNOW, ITS KIND OF PICK YOUR OWN TEXT FOR
20 DEMAGOGUERY AT THIS STAGE. I MIGHT SUGGEST UNDER DB-9% MR.
r
m
BALDUS - DIRECT
1 MCCLESKEY AND A WHOLE LOT OF OTHER FOLKS WOULD BE ON DEATH ROW.
2 AND YOU MIGHT SUGGEST THAT MR. MCCLESKEY MAY OR MAY NOT BE. AND
3 ALL WE’RE DOING IS REALLY GETTING INTO THE TWILIGHT ZONE. I
4 DONT THINK THEY’RE —— THEY“RE INTERESTING, THEY EXPAND THE MIND
S AND I THINK THEYRE PROBABLY GOOD SOCIAL SCIENCE SORT OF WORK
4 PRODUCT, BUT I DON’T THINK THEY PROVE ANYTHING IN THE CASE. AND
7 I WILL NOT ADMIT THEM.
3 | MR. BOGER: YOUR HONOR, IF I COULD ADDRESS THAT FOR A
9 MINUTE.
10 WHAT WE HAD HOPED THEY WOULD INDICATE IS THAT IN THE
11 ABSENCE OF RACIAL FACTORS WE‘D HAVE SOME DIFFERENT SYSTEM THAN
12 WE HAVE NOW, WITH SOME PEOPLE ON DEATH ROW WHO ARE NOT NOW ON
13 DEATH ROW, AND OTHERS NOW ON DEATH ROW WHO ARE NOT.
14 AND QUR SUGGESTION WOULD BE THAT LIKE IN SOME
15 EMPLOYMENT DISCRIMINATION SITUATIONS WHERE THE EFFECT OF A
16 PRACTICE MIGHT NOT HAVE ANY IMPACT, THAT HERE THE EFFECT OF THE
17 RACIAL FACTORS DOES HAVE IMPACT ON WHO ACTUALLY IS ON DEATH ROW
1a AND THE ABSENCE OF RACIAL FACTORS WOULD ALSD AFFECT THE - 1? COMPOSITION OF THAT ROW.
20 AND THAT-S THE SIGNIFICANCE OF THE HYPOTHETICALS.
2} THE COURT: THE SIGNIFICANCE OF THE HYPOTHETICALS ~— IT
22 DOESNT SHOW THAT, THOUGH. AS IT RELATES TO ANY ONE PERSON. IF
23 YOU COMPARE $9 WITH ONE HUNDRED, 99 SUGGESTS THAT THE SAME
24 PEOPLE WOULD BE THERE, PLUS MORE. A HUNDRED SUGGESTS THAT
23 NINETEEN PEOPLE WOULD LEAVE, OR WHATEVER. YOU KNOW, I DON'T
BALDUS - DIRECT
1 | THINK IT”S PROBATIVE OF ANYTHING.
< YOU MAY ARGUE THAT AS A CONCLUSION, THAT BUT FOR THESE
3 FACTORS SOMEBODY MIGHT NOT BE ON DEATH ROW, BUT I DON’T THINK
4 THIS ANALYSIS, WITH THE GREAT TEMPTATION THAT IS PRESENTED
be’ THEREIN. TO USE THAT AS A TEXT FOR ESSENTIALLY DEMAGOGUERY IS
b SUFFICIENTLY RELIABLE FOR ADMISSIBILITY. . |
7 : MR. BOGER: I HOPED. YOUR HONOR, IT WAS GOING TQ FALL A
8 LITTLE SHORT OF DEMAGOGUERY., BUT —-—
2? THE COURT: WELL, THERE‘S A CERTAIN AMOUNT OF BLOODY
0 SHIRT THAT’S IN ALL OF THIS, BUT DEPENDING ON WHICH SIDE OF THE
i1 AISLE YOU’RE ON. YOU CAN TAKE 99 OR ONE HUNDRED AND PROVE
12 ANYTHING YOU WANT TO PROVE. AND I THINK THATS A FAIRLY GOOD
13 INDICATION THAT IT NEITHER PROVES NOR DISPROVES ANY FACT AT
14 ISSUE IN THIS CASE.
15 MR. BOGER: ALL RIGHT.
16 BY MR. BOGER?:
17 Q. PROFESSOR BALDUS., OF COURSE, YOU KNOW THAT GEORGIA HAS AN
18 APPELLATE REVIEW SYSTEM. AND THE APPELLATE REVIEW SYSTEM
po. 12 GOVERNING CAPITAL CASES INCLUDES AN APPELLATE SENTENCE REVIEW.
20 HAVE YOU TAKEN INTO ACCOUNT IN YOUR ANALYSIS THE EFFECT
21 OF THAT REVIEW PROCESS ON THE RACIAL IMPACT THAT WE“VE BEEN
22 TALKING ABOUT?
23 A. YES, I HAVE.
<9 1. HOW DID YOU BEGIN THAT ANALYSIS?
23 A. I BEGAN THE ANALYSIS BY CALCULATING THE PROPORTIONS OF DEATH
or
BALDUS - DIRECT
SENTENCE CASES THAT WERE REVERSED BY DECISION OF THE GEORGIA
SUPREME COURT. I EXAMINED THE GEORGIA REPORTS AND FOUND THAT
GEORGIA SUPREME COURT HAS NOT REVERSED ANY OF THESE CASES ON THE
| GROUNDS THAT RACIAL FACTORS WERE PRESENT IN THEM, OR IN THE
SYSTEM GENERALLY.
2. YOU MEAN EXPRESSLY REVERSED ON THOSE GROUNDS?
A. RIGHT, NONE WAS EXPRESSLY REVERSED ON THOSE GROUNDS.
NEVERTHELESS, THE OEORGIA SUPREME COURT HAS A RECORD OF
REVERSING APPROXIMATELY A QUARTER OF THE DEATH SENTENCE
DECISIONS THAT IT REVIEWS. MANY ON. OR A NUMBER ON BASIC
REVERSALS OF THE CONVICTION, BUT BY AND LARGE BECAUSE OF
REVERSALS OF DECISIONS TAKEN AT PENALTY TRIAL.
AND DB-101 FRESENTS A DISTRIBUTION OF THE REVERSAL
RATES BY THE DEFENDANT-VICTIM RACIAL COMBINATION.
@. WHAT DOES DB-101 REFLECT?
A. IT REFLECTS THAT IN THE CATEGORY OF WHITE VICTIM CASES THAT
THERE“S APPROXIMATELY A 27 PERCENT REVERSAL RATE, AND IN THE
BLACK VICTIM CASES. THAT IT’S APPROXIMATELY 14 PERCENT.
©. DID YOU DO ANY FURTHER ANALYSIS OF THESE UNADJUSTED FIGURES
OF RACIAL -—
A. YES, THE OBJECT OF THE ANALYSIS WAS TO DETERMINE WHETHER THE
IMPACT OF THESE DECISIONS HAD THE EFFECT OF MINIMIZING THE
RACIAL EFFECTS THAT WE HAD OBSERVED IN QUR EARLIER ANALYSIS.
TO MAKE ADJUSTMENTS FOR THAT FACT, 1 REANALYZIED THE
DATA. TREATING EACH DEATH SENTENCE CASE AS A LIFE SENTENCE CASE,
254
BALDUS ~ DIRECT
1 IF THE GEDRGIA SUPREME COURT HAD REVERSED THE DECISION ON
< APPEAL, THAT IS, THE DEATH SENTENCING DECISION ON APFEAL.
|
SO THIS CONSTITUTED A NEW CODING OF THE DEATH SENTENCE (8
DECISION FOR ALL TME CASES, AND THOSE THAT HAD THEIR CASES 4
be’ SUSTAINED. DEATH SENTENCES SUSTAINED WERE TREATED AS DEATH
G-
SENTENCES AND THOSE IN WHICH IT HAS BEEN VACATED WERE TREATED
~4
AS LIFE SENTENCE CAZES.
8. AND THATS YOUR REASON. PERSPECTIVE OF THE REASON FOR REVERSAL?
I A. THATS RIGHT.
10 @. LET ME DIRECT YOUR ATTENTION TO DB-102 FOR IDENTIFICATION
11 AND ASK YOU TO IDENTIFY THAT DOCUMENT?
12 A. ALL RIGHT. DB~-102 IS A TWO-PAGE DOCUMENT. IT IS TABLE 62
13 FROM OUR REPORT. AND IT SHOWS THE REGRESSION CO-EFFICIENTS THAT
14 WE ESTIMATED IN A VARIETY OF DIFFERENT ANALYSES FIRST BEFORE
13 ADJUSTMENT WAS MADE FOR APPELLATE REVIEW. AND THEN AFTER AN
14 ADJUSTMENT WAS MADE FOR APPELLATE REVIEW.
17 AND ONE CAN SEE IN PART 1 OF THIS TABLE THAT WHEN YOU
13 CONSIDER THE BACKGROUND, CORRECTION. THAT WHEN ONE CONSIDERS THE
w 1? DISPARITIES ESTIMATED IN DEATH SENTENCING RATES AMONG OFFENDERS
20 INDICTED FOR MURDER, THAT THERE IS A MODEST DECLINE IN THE
21 MAGNITUDE OF THE RACIAL DISPARITIES THAT ARE OBSERVED AMONG
22 THOSE CASES.
23 FOR EXAMPLE. RACE OF THE VICTIM DISPARITY DROPS FROM .07
<4 TO POINT .04, AND RACE OF THE DEFENDANT DISPARITY DROPS FROM .07
23 TQ POINT .04 AS WELL.
|
|
|
|
BALDUS ~ DIRECT
WHEN, WHEN WE EXAMINE THE JURY DEATH SENTENCING
DECISIONS, THE EFFECT IS GREATER IN THE CHARGING AND SENTENCING
STUDY.
'®%. I’M SORRY, WHERE DO WE LOOK TO EXAMINE THE --
A. I“M SORRY. PART 1B, IN A STEPWISE REGRESSION IN WHICH WE
CONSIDERED FOR POSSIBLE ENTRY INTO THE ANALYSIS ALL OF THE
VARIABLES IN OUR LARGE FILE OF VARIABLES IN THE CHARGING AND
SENTENCING STUDY AND ORIGINAL RACE OF VICTIM EFFECT OBSERVED IN
THE JURY SENTENCING DECISIONS WASHED OUT COMPLETELY.
IN, IN THE PROCEDURAL REFORM STUDY, THOSE RESULTS ARE
REPORTED ON TABLE 2, HERE WE SEE THE, AN ANALYSIS OF THE
COMBINED EFFECTS OF THE PROSECUTORIAL DECISION TO ADVANCE CASES
TO PENALTY TRIAL AS WELL AS THE JURY DECISION AND HERE WE SEE IN
THIS ANALYSIS THAT THE RACE OF VICTIM DISPARITIES ARE NOT
AFFECTED BY THIS ADJUSTMENT FOR BACKGROUND FACTORS, AND THAT IN
THE, FURTHERMORE. IN PART 2B, AT THE BOTTOM, WE SEE THAT THE
RACE OF VICTIM DISPARITIES OBSERVED IN THE JURY SENTENCING
DECISIONS IN THE PROCEDURAL REFORM STUDY ARE REDUCED BY
APPROXIMATELY FIFTY PERCENT.
THE COMBINED IMPRESSION IS, RATHER THE OVERALL
IMPRESSION LEFT BY THIS ANALYSIS IS THAT BECAUSE THE REVERSAL
RATE IS, IN THE GEORGIA SUPREME COURT IS HIGHER IN THE WHITE
VICTIM CASES THAN IT IS IN THE BLACK VICTIM CASES, THAT WE 3EE
AS A CONSEQUENCE OF THOSE APPELLATE DECISIONS SOME REDUCTION IN
THE MAGNITUDE OF THE RACIAL EFFECTS THAT WE HAVE OBSERVED IN THE
v
O
0
~N
>
a
|
|
|
|
PEs
BALDUS - DIRECT
VARIOUS ANALYSES. BUT THAT THEY ARE NOT EXPLAINED AND THAT THEY
REMAIN APPARENT IN THE ANALYSIS.
THE COURT: THEY ARE TEN PERCENTAGE POINTS HIGHER. ON
THE ORDER OF TEN PERCENTAGE POINTS HIGHER.
YOU DON‘T SUGGEST THAT THAT DISPARITY IN RATE AT THE
APPELLATE LEVEL IS AS A RESULT OF ANY RACIAL FACTOR, DO YOU?
THE WITNESS: NO, YOUR HONOR. I DO NOT. OUR ANALYSIS
|BOES NOT SPEAK TO THAT AT ALL.
BY MR. BOCER:
@. WELL. THE EFFECTS THAT YOU SEE IN DB-102, LET ME ASK, DO YOU
TAKE ACCOUNT THERE FOR THE POSSIBILITY THAT SOME OF THE REVERSED
CASES CAN THEMSELVES LATER BE GIVEN A DEATH SENTENCE?
A. OH, YES, THAT, NO. STRIKE THAT.
THIS ANALYSIS DOES NOT TAKE THAT INTO ACCOUNT WHEN WE
LOOK AT THE OVERALL EFFECTS OF THE SYSTEM. THAT IS, TABLE 42,
PART 1, THAT ANALYSIS IS LIMITED TO CASES. THE FIRST, THAT WENT
THROUGH THE SYSTEM FOR THE FIRST TIME.
@. IN OTHER WORDS. IF THE CASE WITH A DEATH SENTENCE WERE
VACATED ON AN EVIDENTIARY PRINCIFLE, WENT BACK TO A SENTENCING
HEARING AND RECEIVED A DEATH SENTENCE AND WAS ULTIMATELY
AFFIRMED, THE ADJUSTMENT REALLY WOULD UNDERESTIMATE THE
CORRECTIVE EFFECT OF THE APPELLATE REVIEW SYSTEM?
A. THAT'S TRUE. AND, IN FACT, OF THE 128 CASES IN OUR STUDY,
THERE ARE TWELVE OF THEM THAT WERE RESENTENCED. ELEVEN. I
BELIEVE, IM NOT SURE, ELEVEN OR TWELVE WERE IN FACT
P
I
I
957
BALDUS — DIRECT
1 RESENTENCED TO DEATH AFTER THE INITIAL VACATION OF THE DEATH
ha
SENTENCE.
3 THE SECTION HERE, 1B, DEALING WITH JURY SENTENCING
4 DECISIONS. ALSO PART 2B, THAT SIMILARLY DEALS WITH JURY
SENTENCING DECISIONS, THAT INCLUDES ALL FENALTY TRIAL CASES. SO
IT DOES NOT HAVE THAT CHARACTERISTIC TO THE SAME DEGREE AS THE
“
d
o
f
e
l
8)
ANALYSIS IN PART 1.
3 MR. BOGER: YOUR HONOR. AT THIS TIME I MOVE THE
9 |ADMISSION OF DB-101 AND 102 INTO EVIDENCE.
10 M3. WESTMORELAND: YOUR HONOR, I DON‘T SEE THE
11 RELEVANCE OF EITHER ONE OF THESE PARTICULAR EXHIBITS AT THIS
12 TIME.
13 MR. BOGER: YOUR HONOR. IF THE STATE WOULD BE WILLING
14 TO STIPULATE THE APPELLATE REVIEW PROCEDURES CANNOT AND DO NOT
1s CORRECT ANY RACIAL DISCRIMINATION THATS FOUND AT THE TRIAL
14 LEVEL, ID BE HAPPY TO WITHDRAW BOTH EXHIBITS.
17 IF THE STATE WON‘T SQ STIPULATE,. THEN WE ANTICIPATE
18 THEY RE GOING TO ARGUE THAT APPELLATE REVIEW MAY HAVE SOME
12? CORRECTIVE EFFECT, IT SEEMS TO ME THIS IS EVIDENCE THATS QUITE
20 RELEVANT.
21 | MS. WESTMORELAND: YOUR HONOR , IT’S AUITE OBVIOUS THAT WE
<2 WOULD NOT S0 STIPULATE, AND I DON‘T SEE THESE EXHIBITS PROBATIVE
23 OF WHAT MR. BOGER HAS ASSERTED AT THIS POINT.
24 MR. BOGER: YOUR HONOR, IF I MAY ADDRESS THE MATTER
23 BRIEFLY? SOMETIMES I HAVE A FEELING THAT WHEN I ADDRESS THE
758
BALDUS - DIRECT
i COURT WHEN IT-S IN THE MIDDLE OF THINKING THAT I“M NOT HELPING
IT AT ALL.
IF THE STATE WILL NOT STIPULATE AS IT SAID THAT THE
APPELLATE REVIEW PROCESS MIGHT NOT CORRECT. THEN IT APPEARS
IMPORTANT FOR US TO SHOW THAT IN FACT THE APPELLATE REVIEW
PROCESS DOESN'T CORRECT FOR ANY RACIAL DISPARITIES THAT EXIST.
THEY DO SO AFTER CONTROLLING FOR 250 VARIABLES. STEPWISE
OH
N
o
p
T
R
d
N
REGRESSION WITH TWO HUNDRED VARIABLES. IN OTHER WORDS, ALL THE
2? KINDS OF CONTROLS THAT THE FEDERAL COURTS LOOK FOR IN THIS KIND
10 OF CASE. IT SEEMS TQ ME IT“S RIGHT ON POINT.
11 1 REALLY DON’T UNDERSTAND. THE STATE MAY WANT TO ARGUE
12 AGAINST THE WEIGHT OF THIS EVIDENCE. BUT ITS PROBITY OR ITS
13 ADMISSIBILITY CERTAINLY SHOULD NOT BE IN GUESTION.
14 THE COURT: I TAKE IT IT’S USEFUL EVERY NOW AND THEN TO
13 TRY TO REMEMBER WHAT ISSUE IT IS YOURE TRYING TO PROVE.
16 WHAT ISSUE ARE YOU TRYING TO PROVE?
17 MR. BOGER: WE“RE TRYING TO PROVE. YOUR HONDR, THAT THE
18 SYSTEM OF CHARGING AND SENTENCING IN THE STATE OF GEORGIA
a i? TOLERATES OR CONTAINS RACIAL FACTORS THAT AFFECT THE OUTCOME OF
<0 SENTENCES IN CAPITAL CASES.
4 FCI THE, BOTH THE FURMAN V. GEDRGIA CASE AND THE GREGG
22 VERSUS GEORGIA CASE SUGGESTED. OR PARTICULARLY THE GREGG
<3 CASE, THAT ONE OF THE FEATURES THAT SAFEGUARDED THE SYSTEM. THAT
24 MADE IT POSSIBLE TO BELIEVE THAT THERE WEREN‘T ARBITRARY OR
23 DISCRIMINATORY FACTORS AT WORK WAS APPELLATE REVIEW.
|
|
259
BALDUS ~ DIRECT
THEREFORE, THE SUPREME COURT, IT APPEARS TO ME, COUNTED
EXPRESSLY ON THE EFFECT OF GEORGIA SUPREME COURT’S SENTENCE
REVIEW IN MITIGATING OR DISSIPATING ANY ABRITRARY FACTORS THAT
MIGHT HAVE APPEARED.
IT SEEMS TO ME THIS TABLE SPEAKS DIRECTLY TO THAT
ASSUMPTION, ASSUMPTION OF WHETHER APPELLATE REVIEW WILL
ELIMINATE ANY FACTOR THAT MAY CREEP IN IN DIFFERENT PARTS OF THE
STATE. IT SUGGESTS THAT THE GEORGIA SUPREME COURTS REVIEW DCES
NOT CORRECT THOSE PROBLEMS, AND THEREFORE —-
THE COURT: ARBITRARY AND CAPRICIOUS LIE UNDER THE
EIGHTH OR 14TH?
MR. BOGER: THAT”S AN 8TH AMENDMENT THEORY. YOUR HONOR.
THE COURT: I WILL ALLOW YOU TQ PROFFER IT IN THAT I
HOPE THIS RECORD WILL DO FOR ALL THE TRIAL JUDGES IN THE
FEDERAL JUDICIARY. IT WILL REMAIN WITH THE RECORD. I DON’T
THINK IT“S ADMISSIBLE FOR ANYTHING THAT IS RELEVANT IN THIS
CIRCUIT, I.E., THAT MCCLESKEY WAS SINGLED QUT BECAUSE oF
IMPERMISSIBLE CRITERIA.
MR. BOGER: ALL RIGHT, YOUR HONOR.
THE COURT: S00. AS TO THOSE DEALING WITH THE EFFECT OF
THE APPELLATE. WHAT IS THAT. 1017
MR. BOGER: 101 AND 102, YOUR HONOR.
THE COURT: THEY ARE MOT ADMITTED, BUT THE CLERK WILL
RETAIN THEM WITH THE PROFFERED EVIDENCE IN A SEPARATE. SEALED
FILE.
260
BALDUS - DIRECT
po
te
MR. BOGER: YOUR HONOR, MIGHT I HAVE A MOMENT TO CONFER
(3)
WITH COUNSEL?
THE COURT: ARE YOU CLOSE TO BEING DONE?
MR. BOGER: VERY CLOSE, YOUR HONOR. WERE THREE OR
FOUR TABLES AWAY FROM FULTON COUNTY.
THE COURT: WE DO GET THERE, THOUGH?
MR. BOGER: WE DO INDEED.
THE COURT: IM TRYING TO RESIST TURNING TO THE LAST
PAGE OF THE NOVEL TO SEE HOW IT ENDS.
o
N
L
y
4
8
d
e
4D
Fo
r MR. BOGER: I“M AFRAID THE LAST PAGE OF THIS NOVEL IS
i1 GOING TO BE OUR PROFFERED REPORT. WHICH HAS COT SO MANY PAGES
12 THAT YOU WON’T COUNT IT A PAGE.
13 BUT I WANT TO CORRECT ONE THING.
14 BY MR. BOGER: |
15 0. 1 WAS ASKING YOU, PROFESSOR BALDUS, ABOUT WHETHER THE EFFECT
16 |OF RESENTENCING ON SOME OF THESE CASES WHERE THERES BEEN A
17 REVERSAL MIGHT LEAVE YOUR DB~102 UNDERESTIMATING THE OVERALL
19 RACIAL EFFECT.
WW 19 | IS THAT YOUR TESTIMONY?
20 A. MY TESTIMONY IS IT COULD HAVE THAT EFFECT WITH RESPECT TO
21 PART 1.
sa, AND IT COULD PART 1A. BECAUSE THAT DOES NOT TAKE ANY
23 ACCOUNT OF SUBSEQUENT JURY DECISIONS AT RETRIAL, WHEREAS PART 1B
24 DOES INCLUDE SUBSEQUENT RETRIALS FOR THOSE CASES THAT HAVE HAD
23 TIME TO COME BACK INTO THE SYSTEM AND FALL WITHIN THE FRAMEWORK
P61
BALDUS ~ DIRECT
OF NUR STUDY.
IT MAY ALSO BE THAT IT MAY UNDERCOUNT THEM TO THE
EXTENT THAT SOME OF THE LATER CASES THAT ARE INCLUDED IN THE
|
ANALYSIS IN 1B ARE VACATED AND COME BACK IN AND RECEIVE DEATH
SENTENCES WHICH THEMSELVES ARE NOT VACATED AT A LATER TIME.
BUT I THINK THE UNDERESTIMATION OF THOSE EFFECTS OF
REVERSALS IS GREATER TO THE EXTENT IT EXISTS IN PART 1A THAN IT
DOES IN PART 1B. BECAUSE WE HAVE MANY OF THE RETRIAL CASES IN
THE 1B.
D. ALL RIGHT. THANK YOU.
NOW,» DID YOU CONDUCT ANY ANALYSIS TO SEE WHETHER THE
EFFECTS OF RACIAL DISPARITIES THAT YOU HAD UNCOVERED MIGHT BE A
FUNCTION OF A PARTICULAR SUB-PERIOD WITHIN THE GROUP OF YEARS
DURING WHICH YOUR STUDY WAS CONDUCTED?
A. YES, 1 DID.
9. LET ME ASK YOU TO TURN TO DB-103 AND IDENTIFY THAT DOCUMENT?
‘A. DB-103 IS A. IS TABLE &3 FROM THE CHARGING AND SENTENCING
STUDY.
@. WHAT DOES IT REFLECT?
A. IT REFLECTS ANALYSES OF DATA FROM PARTICULAR TIME PERIODS
| DURING THE TIME FRAME OF OUR STUDY.
THERES A OROWING BODY OF LITERATURE IN THE AREA OF
{
i
[
SENTENCING WHICH FOCUSES ON THE QUESTION OF TEMPORAL CHANGES IN
THE IMPACT OF ILLEGITIMATE AND SUSPECT FACTORS IN SENTENCING
SYSTEMS. IT IS KNOWN THAT THERE ARE MANY FACTORS INFLUENCING
ee amas Soo—_——————p———" —— TE — L.A — S— S————————. TYE | A ——— ——————. S———_ T——_ JE S——— S— ST ———_" TE - SR m—r— — —
BALLS - DIRECT |
|
c MANGES IN SYSTEMS. CHANGING IN ALTITURES., CHANGES IN PERSONNEL,
2 THERE ARE MANY FACTORS THAT INFLUENCE A SYSTEM. AND BEFORE ONE
can MAKE ANY ASSESSMENT ABOUT THE SYSTEM AS A WHOLE IT’S
IMPORTANT TO DISAGGREGATE THE DATA AND LOOK AT DURING DISCREET
TIME PERIODS.
HAL S WHEAT WE DID HERE. WE COMPARE HERE THE
REGRESSION CO-EFFICIENTS FOR RACE OF VICTIM AND RACE OF
DEFENDANT ESTIMATED WITH WEICHTED LEAST SQUARES. FIRST. AS THE
wv
OO
«N
N
OC
4a
P
w
BENCH MARK FOR COMPARISON, ARE THE CO-EFFICIENT ESTIMATED FOR
10 THE ENTIRE PERIOD. AND THEN WE BROKE THE DATA DOWN INTO TWO-YEAR
11 PERIODS, AND ESTIMATED CO-EFFICIENTS FOR THEM.
12 'Q. THESE ARE UNADJUSTED OR ADJUSTED?
13 A. THESE ARE ADJUSTED CO-EFFICIENTS.
34 THE COURT: ADJUSTED FOR WHAT?
15 THE WITNESS: ADJUSTED FOR THE. IN A STEPWISE
14 REGRESSION PROCEDURE, YOUR HONOR, THAT SCREENS THROUGH THE
17 ENTIRE FILE OF OVER TWO HUNDRED AND SOME VARIABLES. MOST OF 13 THESE ANALYSES ENDED UP WITH FROM 4Q TO, 30 TO FIFTY VARIABLES
w 1? IN THEM. THE SAME TYPE OF PROCEDURE, STEPWISE PROCEDURE THAT
20 WAS USED IN THE PRINCIPAL PART OF THE ANALYSIS EARLIER. YOUR
23 HONOR.
ae AND THE RESULT SHOWS THAT THERE IS A CERTAIN AMOUNT OF
(INSTABILITY IN THE RACIAL EFFECTS THAT ARE OBSERVED OVER THIS,
|
24 THESE DIFFERENT TIME PERIODS, AND THAT. BUT THAT THERE IS NO
23 STRONG TREND, I GUESS WOULD BE THE POINT, IT EMERGES MOST
ot
SN
R
1
E
E
4
8
@
1Q
263
BALDUS - DIRECT
CLEARLY. THERES NO TREND EITHER UPWARD OR DOWNWARD IN THESE
DATA. IT SUGGESTS THAT THESE FACTORS, THE PROPERTIES OF THE
SYSTEM THAT ARE PRODUCING THESE EFFECTS APPEAR TO HAVE BEEN
WORKING IN ONE DEGREE OR ANOTHER THROUGHOUT THE PERIOD COVERED
BY THE STUDY.
MR. BOGER: YDUR HONOR, AT THIS TIME I MOVE THE
ADMISSION OF DB-103 INTO EVIDENCE.
MS. WESTMORELAND: YOUR HONOR, I FAIL TO SEE THE
RELEVANCE OF THIS PARTICULAR EXHIBIT. ONCE AGAIN WE'RE
CONCERNED WITH THIS PARTICULAR CASE AND NOT WHAT MAY HAVE TAKEN
PLACE IN 1974, “78 OR ANY PARTICULAR YEAR.
THE COURT: IT-S A BOOTSTRAPPING OR DATA VALIDATING OR
WHATEVER ELSE HE WANTS TO CALL IT. I DON'T THINK IT PROVES ANY
FACT IN ISSUE, INSOMUCH AS IT ADDRESSES THE RELIABILITY OF THE STUDY. AND TO THE EXTENT THAT ANYTHING WE-“VE HEARD ABOUT THE
RELIABILITY OF THE STUDY IS ADMISSIBLE, I THINK THAT IT IS FOR
ret REASON.
MR. BOGER: THANK YOu, YOUR HONOR.
I MIGHT ADD THAT I CERTAINLY FEAR IF WE HAD NOT
INCLUDED SUCH A CHART, THE STATE WOULD WELL HAVE SAID, --
| THE COURT: I“M SURE THEY WOULD, IF THEY THOUGHT ABOUT
IT.
BY MR. BOGER:
@. PROFESSOR BALDUS, DID YOU CONDUCT ANY ANALYSIS THAT LOOKED
AT DIFFERENTIAL TREATMENT IN URBAN AND RURAL AREAS WITHIN THE
]
r
0
m
N
o
e
r
4
764
BALDUS ~ DIRECT
STATE OF GEORGIA. AS A POSSIBLE EXPLANATION FOR THE RACIAL
DISPARITIES THAT MIGHT EXIST IN THE SYSTEM?
'A. YES. WE DID.
Q@. LET ME ASK YOU TO DIRECT YOUR ATTENTION TO DB-104 MARKED FOR
IDENTIFICATION, ASK YOU TO IDENTIFY THAT DOCUMENT?
A. DB-104 IS TABLE 6& FROM OUR REPORT. AND IT PRESENTS URBAN
AND RURAL RESULTS PRODUCED IN A SERIES OF DIFFERENT ANALYSES.
Q. NOW WHY WERE YOU INTERESTED IN A RURAL~URBAN BREAKDOWN?
A. WELL, BASICALLY, OUR CONCERN WAS SOMEWHAT BROADER THAN
SIMPLY THE RURAL-URBAN BREAKDOWN. WE WERE INTERESTED IN TESTING
THE PLAUSIBILITY OF A RIVAL HYPOTHESIS. JUST AS IN THE LAST
ANALYSIS A PLAUSIBLE IS THE RACIAL DISPARITIES THAT YOU SEE ARE
CONCENTRATED IN A PARTICULAR PERIOD OF TIME AND THEREFORE THEY
THREATEN THE VALIDITY OF ANY ASSERTION THERES A SYSTEM-WIDE
EFFECT.
SIMILARLY, THERE IS A PLAUSIBLE HYPOTHESIS THAT ONE
NEEDS TO TEST IN THIS KIND OF RESEARCH. AND IT IS AS FOLLOWS.
THIS IS A DECISION MAKING SYSTEM THATS HIGHLY
DECENTRALIZED. DECISIONS ARE MADE IN COUNTIES. BUT THE
PRINCIPAL ADMINISTRATIVE OFFICER WHO HAS INFLUENCE IN THIS
SYSTEM IS THE PROSECUTOR. THE PROSECUTORS OPERATE ON A
DISTRICT- OR CIRCUIT~WIDE LEVEL WITHIN THE STATE JUDICIAL
CIRCUIT, OF WHICH THERE ARE FORTY-TWO.
NOW IT IS ENTIRELY PLAUSIBLE THAT. THAT IN FACT THE
SYSTEM OPERATES IN A PERFECTLY EVENHANDED FASHION WITH RESPECT
P65
BALDUS - DIRECT
1 TO THESE FASHION FACTORS. RACIAL FACTORS. BUT THERE ARE CERTAIN
2 CIRCUITS WHERE PROSECUTORS AND JURIES ARE HEAVILY INCLINED TO
3 IMPOSE DEATH SENTENCES AND DEATH SENTENCES ARE IMPOSED AT
4 SIGNIFICANTLY HIGHER RATES IN THOSE CIRCUITS THAN OTHER
= CIRCUITS. IN THOSE CIRCUITS. EVERYONE IS TREATED IN AN
b EVENHANDED FASHION, AND FURTHERMORE, THOSE ARE THE CIRCUITS
4 WHERE YOU FIND MOST OF THE WHITE VICTIM CASES. WHITE VICTIM
a CASES ARE HEAVILY CONCENTRATED IN THOSE CIRCUITS, AND THAT WOULD
? HAVE TO AFFECT, WHEN YOU LOOK AT THE DATA STATEWIDE, OF
10 SUGGESTING RACE OF VICTIM EFFECTS THROUGHOUT THE JURISDICTION.
11 WHEN IN FACT. THERE IS AN EVENHANDED TREATMENT WITHIN THE
12 | SYSTEM. 13 | THERE“S A VERY IMPORTANT ARTICLE IN THE LITERATURE THAT
14 ILLUSTRATES THIS PHENOMENON. IT“S KNOWN AS SIMPSON’S PARADOX.
13 AND ITS ILLUSTRATED BY A VERY INTERESTING STUDY THAT SOMEONE
14 DID.
17 SOMEONE WAS CLAIMING THAT BERKELEY UNIVERSITY WAS
18 DISCRIMINATING AGAINST FEMALES IN ADMISSION TO GRADUATE SCHOOL.
ww 19 INDEED WHEN YOU LOOK AT THE AGGREGATE ADMISSION DATA IN BERKELEY 20 UNIVERSITY. THAT WOMEN WERE ADMITTED AT A MUCH LOWER RATE THAN
21 MEN AND IT WAS ON THE BASIS OF THOSE DATA WHICH COVERED THE
22 ENTIRE SYSTEM THE CLAIM OF SEX DISCRIMINATION WAS ADVANCED.
23 BERKELEY UNIVERSITY OPERATES LIKE THE GEORGIA CHARGING
24 AND SENTENCING SYSTEM. IT-5S HIGHLY DECENTRALIZED INTO THE
23 DEPARTMENTS WITHIN THE UNIVERSITY.
| Db
BALDUS ~ DIRECT
1 | AN INVESTIGATION OF THE DATA AT THE DEPARTMENTAL LEVEL
2 |SHOWED THAT THE REASON FOR THE OVERALL DISPARITIES WAS THAT
3 EN PRINCIPALLY APPLIED TO THE DEPARTMENTS THAT HAD VERY LOW
4 |OVERALL SELECTION RATES. EVERYBODY WAS BEING TREATED EQUALLY,
5 FOR EXAMPLE, IN THE ENGLISH DEPARTMENT. BUT THERE WERE
& PRINCIPALLY WOMEN APPLICANTS AND THE SELECTION RATE FOR PEOPLE
7 IN THAT DEPARTMENT WAS EXTREMELY LOW. THE STANDARDS WERE VERY
& (HIGH. SO MOST OF THE WOMEN APPLYING THERE WERE REJECTED WHEREAS
9 IN THE ENGINEERING DEPARTMENT, FOR EXAMPLE. THE ACCEPTANCE RATES
10 WERE VERY HIGH. THOSE DEPARTMENTS OF THAT TYPE, WHERE THERE
11 WERE HIGH ACCEPTANCE RATES. HAD PRINCIPALLY MALE APPLICANTS. AND
iz IN BOTH OF THOSE DEPARTMENTS, IN FACT, THE WOMEN TENDED TO BE
13 |FAVDRED SLIGHTLY OVER THE MEN, BOTH IN THE ENGINEERING
14 [DEPARTMENT WHERE THE RATES WERE HIGH, AND ALSO IN THE HARD
15 DEPARTMENTS WHERE THE HIGHER STANDARDS WERE IMPOSED, WHERE THE
14 WOMEN WERE PRINCIPALLY APPLYING.
17 S0 THE CONSEQUENCE WAS, WHEN YOU AGCGREGATE THE DATA
18 TOGETHER, THE OVERALL EFFECT SUGGESTS THAT WOMEN ARE BEING
Ww 19 DISADVANTAGED IN THE SYSTEM, BUT WHEN YOU GO DOWN AND LOOK IN
20 THE INDIVIDUAL DECISION MAKING UNITS, WHERE DISCRETION IS
21 ACTUALLY BEING EXERCISED, YOU FOUND JUST THE CONTRARY, NOT
22 STRONGLY BUT CERTAINLY THERE WAS NQ EVIDENCE OF ANY.BIAS AGAINST
23 WOMEN.
24 S0 THAT IS ALWAYS A POTENTIAL RIVAL HYPOTHESIS WHENEVER
23 YOU’RE DEALING WITH A HIGHLY DECENTRALIZED SYSTEM OF THE TYPE
fy
0
N
O
967
BALDUS ~- DIRECT
WERE CONCERNED WITH HERE IN THE STATE OF GEORGIA.
SO, TO. TO TEST THE PLAUSIBILITY OF THAT HYPOTHESIS, WE
EXAMINED THE DEATH SENTENCING RATES IN THE VARIOUS JUDICIAL
CIRCUITS.
WE DID IT AS WELL IN THE COUNTIES, BUT THE COUNTIES
HAVE, MOST OF THE COUNTIES HAVE SO FEW CASES IT“S VERY HARD Ta
MAKE ANY SORT OF JUDGMENT, SO WE PRINCIPALLY CONCENTRATED THE
ANALYSIS IN THE JUDICIAL CIRCUITS.
AND WITHOUT ANY ADJUSTMENT FOR BACKGROUND FACTORS. WE
LOOKED AT THE DEATH SENTENCING RATES AMONG THE JUDICIAL
CIRCUITS. AND DID SIMPLE CROSS TABULATIONS WHICH RELATED THE
PROPORTION OF WHITE VICTIM CASES IN THOSE CIRCUITS WITH THE
DEATH SENTENCING RATES IN THOSE CIRCUITS.
WE DIDN’T FIND ANY EVIDENCE THAT WHITE VICTIM CASES
WERE HIGHLY CONCENTRATED IN THE CIRCUITS WHERE THE HIGHEST RATES
WERE, THERE WAS NOTHING TO SUPPORT THAT.
WE ALSO FOUND THAT IN THE CIRCUITS THAT HAD, THAT. AT
THIS LEVEL OF ANALYSIS, THERE WERE DISPARITIES IN TERMS OF THE
RATES. AT THE CIRCUIT LEVEL. BUT A MORE PROBATIVE WAY TO LOOK
AT THAT IS TO AGGREGATE THE DATA. THATS WHY WE DID THE
URBAN-RURAL ANALYSIS BECAUSE IT MAY VERY WELL BE THERE ARE
SHARFER DIFFERENCES IN TERMS OF DEATH SENTENCING RATES IN
URBAN-RURAL SETTINGS IN GEORGIA, IN FACT, THERE ARE. THE DEATH
SENTENCING RATE IS SUBSTANTIALLY HIGHER IN THE RURAL AREAS THAN
IN THE URBAN AREAS OF THE STATE. AND WE MIGHT SEE SOME EVIDENCE
F683
BALDUS ~ DIRECT
OF SIMPSON-3 PARADOX IN THAT CONTEXT. THAT WAS THE REASON WE
MADE THIS COMPARISON BETWEEN THE URBAN AND THE RURAL CIRCUITS.
AND WHEN WE DID $0, WE FOUND THE EVIDENCE THAT IS
REPORTED HERE IN DB-104, AND IT SUGGESTS THAT. THAT WE FIND RACE
OF VICTIM DISPARITIES ARE NOT CONCENTRATED STRICTLY IN ONE
PARTICULAR AREA OF THE STATE, BUT THAT THERE ARE RACE OF VICTIM
EFFECTS APPARENT IN BOTH URBAN AND RURAL AREAS, ALTHOUGH THEY
ARE LARGER AND MORE STATISTICALLY SIGNIFICANT IN RURAL AREAS.
AND WE FIND THAT THE RACE OF DEFENDANT EFFECTS ARE
FRINCIFPALLY A PRODUCT OF DECISION MAKING IN RURAL AREAS.
THAT CAN BE SEEN MORE CLEARLY IF ONE WILL LOOK AT PART
2 OF DB-104, HERE WE SEE A COMPARISON AMONG THE. THE
APPROXIMATELY FIVE HUNDRED CASES WITH THE HIGHEST RISK OF
RECEIVING A DEATH SENTENCE, WHERE I BELIEVE THERE’S THE GREATEST
ROOM FOR THE EXERCISE OF DISCRETION, THIS IS THE POPULATION
WHERE I THINK YOU WOULD HAVE THE MOST. THE GREATEST LIKELIHOOD
OF SEEING A RACIAL EFFECT.
AND WHEN WE LOOK INTO THIS POOL OF CASES AND SUBDIVIDE
1T BETWEEN URBAN AND RURAL AREAS, WE SEE THE RESULTS IN PART 2
OF DB~104. ROW A OF PART 2, SHOWS THAT THE RACE OF VICTIM
EFFECT IS SUBSTANTIALLY LARGER IN THE RURAL AREAS. IT HAS A
HIGHER LEVEL. HIGHER DEGREE OF STATISTICAL SIGNIFICANCE.
WHEN YOU LOOK AT THE NEXT COLUMN OVER. WHERE IT SAYS
RACE OF DEFENDANT, ALL CASES, YOU MAKE A COMPARISON. YOU SEE
THAT THE DISPARITY IS MUCH LARGER IN THE RURAL AREAS.
6?
BALDUS - DIRECT
Pr
os
t AND THEN WHEN YOU FOCUS IN ON THE WHITE VICTIM CASES
ONLY. AND AS WE HAVE BEEN SUGGESTING HERE, THE EVIDENCE SUGGESTS IX
3 THAT RACE OF DEFENDANT EFFECTS ARE PRINCIPALLY THE PRODUCT OF
4 |THE PROCESSING OF WHITE VICTIM CASES, YOU SEE HERE A VERY SHARP
3 DIFFERENCE IN TERMS OF THE URBAN AND RURAL SCENE. IN THE URBAN
o
n
SCENE THERE'S A VERY MODEST IMPACT THAT'S NOT STATISTICALLY
® .
7 SIGNIFICANT WHEREAS IN THE RURAL CONTEXT YOU SEE A VERY
a SUBSTANTIAL EFFECT WHICH IS HIGHLY SIGNIFICANT FROM A
bs STATISTICAL STANDPOINT.
10 THE COURT: ONE -- ARE YOU DONE?
11 THE WITNESS: YES, YOUR HONOR,
12 THE COURT: ONE OF THE THINGS THAT BOWERS AND PIERCE
13 DID THAT I THOUGHT WAS INTERESTING, AND I GUESS THERE’S A
14 CERTAIN AMOUNT OF INTUITIVE VALIDATION OF THIS, THEY DID A STUDY
13 ON REGIONAL, SUBREGIONAL VARIATIONS, AND FOUND THAT THERE WAS A
146 SUBSTANTIAL DIFFERENCE IN SENTENCING RATES, IMPOSITION RATE,
17 WITHOUT REFERENCE TO RACE PERHAPS. BETWEEN SQUTH CEORGIA. MIDDLE
13 GEORGIA, NORTH GEORGIA AND ATLANTA. AND THERE ARE SOME. THERE
w iP 15 SOME RATIONALE FOR THAT SORT OF A STRATIFICATION. 20 DID YOU MAKE A BREAKDOWN IN THE RURAL BEYOND JUST RURAL
21 |AND URBAN, AND WHAT IS YOUR DEFINITION OF RURAL AND WHAT I3 YOUR
2 (DEFINITION OF URBAN?
23 | THE WITNESS: THE, THE DEFINITION OF URBAN USED HERE,
248 YOUR HONOR, IS THAT MORE THAN FIFTY PERCENT OF THE POPULATION
23 LIVES IN AN URBAN PLACE AS DEFINED BY THE BUREAU OF THE CENSUS,
2
WO
MN
Le
| 270
BALDUS - DIRECT
THE WAY WE SPECIFIED IT. OVER FIFTY PERCENT OF THE PEOPLE IN
|THE CIRCUIT LIVE IN AN AREA, A PLACE DEFINED BY THE BUREAU AS
HRGA.
THE COURT: YOU'RE TALKING ABOUT STANDARD METROPOLITAN
AREA WOULD BE, ANY COUNTY IN A STANDARD METROPOLITAN AREA WOULD
BE CLASSIFIED AS URBAN?
THE WITNESS: I DON‘T KNOW THE ANSWER TO THAT, YOUR
HONOR, WHETHER IT WOULD OR NOT. WE WERE GUIDED BY THE CENSUS
BUREAUS DEFINITION OF URBAN. CERTAINLY ALL THIS AREA HERE,
ATLANTA IS URBAN.
THE COURT: ANYBODY WHO KNOWS ANYTHING ABOUT ATLANTA
WILL TELL YOU THAT THE TRUE URBAN AREA IS PROBABLY A TWO-COUNTY
AREA, AND UNDER THE GREATEST STRETCH OF THE IMAGINATION IS NO
GREATER THAN A FIVE-COUNTY AREA.
BUT THE CENSUS FOLKS THINK IT’S A 31-COUNTY AREA. AND
ANY SIMILARITY BETWEEN THE ATTITUDES IN FAYETTE COUNTY ARE
PURELY COINCIDENTAL AND NEITHER ONE OF THOSE WOULD COINCIDE WITH
THE ATTITUDES IN THE ATLANTA JUDICIAL CIRCUIT. I THINK I
COULD ALMOST JUDICIALLY NOTICE THAT.
THE WITNESS: WELL, WHAT WE DID, YOUR HONOR. WE DID
REPLICATE BOWERS AND PIERCE’S ANALYSIS, USING THE FIVE REGIONS
"OF THE STATE. WE BROKE DOWN THE AREA INTO THE REGIONS THEY DID.
WE TRACKED THEIR CIRCUIT-WIDE BREAKDOWN.
THE COURT: DOES THAT COME UP OVER HERE A LITTLE BIT?
THE WITNESS: NO, IT IS NOT INCLUDED IN THIS REPORT.
r
4]
|
71
BALDUS - DIRECT
THE COURT: WELL, YOU KNOW, THAT WAS INTERESTING TO ME
BECAUSE IF I REMEMBER THE FIGURES, WHAT THEY DEFINED I THINK AS
MIDDLE GEORGIA HAD THE HIGHEST INCIDENCE OF IMPOSITIONS WHAT
THEY CLASSIFIED AS SOUTH GEORGIA HAD THE SECOND HIGHEST: WHAT
THEY CLASSIFIED AS ATLANTA HAD THE NEXT HIGHEST? AND WHAT THEY
CLASSIFIED AS NORTH GEORGIA HAD THE NEXT HIGHEST, WHICH WOULD
SUGGEST THAT WITHIN RURAL, THERE I5 A SUBSTANTIAL DIFFERENCE IN
ATTITUDE OF SOMEBODY, PROSECUTORS, JURY, SOMEBODY IN THE SYSTEM.
AND I THINK A LOT OF PEOPLE WHO GREW UP IN GEORGIA PROBABLY
THINK THAT THERE ARE A LOT OF DIFFERENCES IN ATTITUDE AS PERHAFS
BETWEEN EAST AND WEST IOWA. NORTH AND SOUTH IOWA, WHATEVER.
THATS INTERESTING.
ALL RIGHT.
THE WITNESS: I JUST WANTED TO MENTION ONE OTHER
ANALYSIS THAT WAS DONE, YOUR HONOR, TO TRY AND DEAL WITH THIS
POTENTIAL PROBLEM WHERE WE’RE SEEING THE EFFECTS OF HIGH DEATH
SENTENCING RATES IN PARTICULAR JURISDICTIONS WHERE THERE ARE A
LOT OF WHITE VICTIMS.
ANOTHER WAY OF DEALING WITH THIS IS TO INTRODUCE INTO A
REGRESSION ANALYSIS A VARIABLE THAT STANDS FOR EACH COUNTY, AND
1
}
WE DID THAT. WE DID REGRESSION ANALYSES CONTROLLING FOR LEGITIMATE FACTORS BUT ALSO INCLUDING IN THE ANALYSIS FORTY-TWO
VARIABLES. EACH ONE STANDING FOR A CIRCUIT IN THE STATE. AND
WHAT THAT HAS THE EFFECT OF DOING IS ALLOWING THE EXPLANATORY
POWER OF THESE CIRCUITS WHICH HAVE HIGH DEATH SENTENCING RATES,
272
BALDUS - DIRECT
3 DEATH SENTENCING RATES THAT ARE ABOVE AVERAGE. TO EXPLAIN THE
2 VARIATIONS IN DEATH SENTENCING RATES, AND THIS PHENOMENON THAT
3 '1“M DESCRIBING WERE REALLY AT WORK, THAT WOULD TEND TO ELIMINATE
4 THE RACIAL EFFECTS, AND WHEN WE DID THAT SORT OF ANALYSIS, IT
3 DID NOT HAVE ANY EFFECT ON THE RACIAL EFFECT. THAT WAS JUST
A & ANOTHER WAY OF TESTING THE HYPOTHESIS THAT WHAT WE MIGHT BE
7 SEEING ARE THE EFFECTS OF SIMPSON’S PARADOX.
8 MR. BOGER: YOUR HONOR, AT THIS TIME, I WOULD MOVE THE
? ADMISSION OF DB-104 INTO EVIDENCE.
10 MS. WESTMORELAND: YOUR HONOR, ILL OBJECT TO DB-104
11 FOR THE SAME REASONS PREVIOUSLY GIVEN AND ALSO ADDITIONALLY I
12 WOULD OBJECT, I DON‘T SEE ITS RELEVANT TO THE DISCUSSION AT
13 HAND, PARTICULARLY IN THE LIGHT OF THE DEFINITIONS UTILIZED FOR
14 URBAN AND RURAL.
| THE COURT: I AM TROUBLED BY THE LACK OF KNOWLEDGE AS
16 TO WHAT URBAN MEANS AND WHAT RURAL IS, BUT I WILL ADMIT IT.
17 MR. BOGER: THANK YOU, YOUR HONOR,
18 BY MR. BOGER:
w 19 Q. FINALLY, BEFORE WE REACH FULTON COUNTY, PROFESSOR BALDUS,
20 yoy HAD TESTIFIED EARLIER THAT A NUMBER OF DEFENDANTS IN THE
21 CAPITAL SYSTEM HAD PLEADED GUILTY TD MURDER. AND THEN RATHER
22 THAN BINDING THEMSELVES TO LIFE SENTENCE, HAD GONE ON TO PENALTY
23 TRIALS, JUDGE TRIALS, AT WHICH THE JUDGE IMPOSED THE SENTENCE OF
<4 LIFE OR DEATH. 23 DID YOU DO ANY ANALYSIS TO IDENTIFY THE POSSIBLE
ro
ow
it
h
o
n
R
U
A
4
BR
EE
p
e
1
23
73
BALDUS - DIRECT
| SENTENCING OUTCOMES IF ONLY JURY SENTENCING DECISIONS WERE
CONSIDERED AND THE JUDGE SENTENCINGS WERE OMITTED?
‘0. LET ME DIRECT YOU TO DB-105 MARKED FOR IDENTIFICATION AND
AND I WILL ASK YOU TO IDENTIFY THAT DOCUMENT?
A. YES, DB-~10%, WHICH IS TABLE 6&7 FROM OUR REPORT, PRESENTS THE
RESULTS OF ANALYSIS CONDUCTED WITH DATA FROM THE CHARGING AND
SENTENCING STUDY. |
THE NUMBER OF CASES IN WHICH A DEFENDANT IS,
DEFENDANTS CASE IS ADVANCED TO A PENALTY TRIAL AFTER THE ENTRY
OF A GUILTY PLEA IN A MURDER CASE IS VERY SMALL, REPRESENTS LESS
THAN FIVE PERCENT OF THE CASES, MURDER CASES, RATHER PENALTY
{TRIAL CASES BY MY ESTIMATE, AND NEVERTHELESS WE FELT IT
IMPORTANT TO CONTROL FOR THE POSSIBILITY THAT IT WAS JUDICIAL
DECISIONS THAT MIGHT BE PRODUCING THIS EFFECT, RATHER THAN JURY
DECISIONS OR VICE VERSA. WE THOUGHT IT WAS IMPORTANT TO ATTEMPT
TO DISENTANGLE THE RESPECTIVE ROLES OF JUDGE AND JURY IN THE
JURISDICTION.
S0, WE CONDUCTED ONE ANALYSIS INCLUDING ALL OF THE
PENALTY TRIAL CASES AND THEN AN ALTERNATIVE ANALYSIS RESTRICTING
THE DATA TO CASES IN WHICH THERE WAS A JURY PENALTY TRIAL
DECISION, AND THE RESULTS ARE PRESENTED HERE IN DB-103.
2. WHAT DO THEY REFLECT?
A. THEY REFLECT THAT AT THE. THEY REFLECT OVERALL THAT IT MAKES
NO DIFFERENCE IN THE ANALYSIS, THAT WHEN YOU LOOK AT THE
974
BALDUS - DIRECT
po
ey
DISPARITIES ESTIMATED AMONG ALL DEFENDANTS INDICTED FOR MURDER,
3 THAT YOU SEE NO EFFECT. AND WITH RESPECT TO THE RACE OF THE
VICTIM, JUST A TRACE OF EFFECT WITH RESPECT TO THE RACE OF
DEFENDANT, AND THAT WITH RESPECT TO THE DEATH SENTENCING
DECISION OF THE JURY, YOU SEE VERY LITTLE CHANGE AS WELL.
OUR CONCLUSION IS THAT BECAUSE THE BENCH PLAYS SUCH A
SMALL ROLE IN THIS PROCESS OF SENTENCING, THAT THE, IT HAS NO
REAL EFFECT ON IT. AND THAT THE ANALYSES THAT WE HAVE CONDUCTED
THAT INCLUDE BOTH JURY AND BENCH DECISIONS ARE BASICALLY
l
e
B
T
R
TT
i
REPRESENTATIVE OF JURY BEHAVIOR.
11 MR. BOGER: YOUR HONOR, AT THIS TIME I WILL MOVE THE
12 ADMISSION OF DB-10S INTO EVIDENCE.
13 MS. WESTMORELAND: YOUR HONOR, I OBJECT. I FAIL TO SEE
14 | THE RELEVANCE TO THIS CASE. THE INSTANT CASE IS CLEARLY NOT ONE
13 IN WHICH A JUDGE MADE ANY SENTENCING DECISION. THIS IS JURY
16 SENTENCING CASE.
37 MR. BOGER: IN A SENSE I THINK THAT 3 THE POINT, YOUR
13 HONOR. WE WANTED TO DEMONSTRATE THAT.
W 1%? THE COURT: ILL ADMIT IT.
20 WE“LL TAKE A TEN-MINUTE BREAK UNTIL ABOUT 3:20.
3 MR. BOGER: THANK YOU,
22 J" yo
23 (RECESS TAKEN.?
24 MS. BOLEYN: YOUR HONOR, IF THE COURT PLEASE, AND MR.
29 BOGER. I°VE ALREADY ASKED HIS PERMISSION, IF 1 COULD BRING ONE
975
BALDUS ~ DIRECT
1 MATTER TO THE COURT’S ATTENTION?
< AS THE COURT KNOWS, THE ATTORNEY GENERALS OFFICE ALSO
3 | REPRESENTS THE STATE BOARD OF PARDONS AND PAROLES AND THIS
4 MORNING AT 10130 A SUBPOENA WAS SERVED ON MR. WARR, W-A-R-R, OF
3 THE STATE BOARD OF PARDONS AND PAROLES BY MR. STROUP ACTING FOR
o
>
THE PETITIONER, SUBPOENA DIRECTED HIM TO APPEAR AT TWO 0-CLOCK
7 | THIS AFTERNOON. |
3 AND HAVING BEEN HERE THIS MORNING AND HEARD THE COURTS
9 |COMMENTS ABOUT THE AGENDA FOR TODAY, I ASKED MR. WARR TO REMAIN
10 |IN OFFICE AND BE ABLE TO BE ON CALL AND TO BE ABLE TO BE HERE IN
11 |CASE MR. BOGER OR STROUP WANTED HIM. HE IS AVAILABLE THIS
12 |AFTERNDON, BUT AT SOME POINT I°D LIKE TO SEEK THE OUIDANCE OF
13 |THE COURT AND THE COUNSEL FOR THE PETITIONER TO KNOW WHETHER
14 |OR NOT WE CAN RELEASE HIM. HE IS WAITING IN HIS OFFICE PURSUANT
1% |TO THE SUBPOENA AND WE CAN HAVE HIM OVER HERE AT ANY TIME. 1
16 |WAS JUST UNCERTAIN AS TO WHETHER OR NOT HE WOULD BE NEEDED TODAY
i7 AT ALL. 18 MR. STROUP: YOUR HONOR, I CAN SPEAK TO THAT. I HAVE
Eo 19 TRIED TO BE IN TOUCH WITH MR. WARR AND HAVE BEEN UNABLE TO REACH
20 HIM.
WE HAD THOUGHT THERE MIGHT BE SOME POSSIBILITY OF
§
|
|
REACHING HIM THIS AFTERNOON. THAT WAS AT LEAST OUR READING
22 EARLY THIS MORNING. AND I HAVE TRIED TO REACH HIM TO ADVISE HIM
24 SIMPLY TO BE ON CALL FOR MONDAY, BUT HAVE BEEN UNABLE TQ DO
25 S0.
SH
W
M
«a
— | —r—— ——— ————— p———— ——
P76
BALDUS ~ DIRECT
THE COURT: YOU ARE GOING TO CROSS PROFESSOR BALDUS?
MS. WESTMORELAND: WHEN MR. BOGER FINISHES, YES, YOUR
Joncn.
THE COURT: YOU WANT THIS FELLOW FROM FARDONS AND
PAROLES AFTER PROFESSOR BALDUS AND BEFORE THE STATISTICIAN OR
AFTER THE STATISTICIAN?
| MR. BOGER: WELL, YOUR HONOR, I THINK OUR THOUGHT ON
THIS WAS ONE OF NOT KNOWING EXACTLY WHEN WE WOULD BE FINISHED
WITH PROFESSOR BALDUS AT THE OUTSET OF THE DAY, AND NOT KNOWING
HOW LONG THEY WERE GOING TO TAKE ON CROSS. BUT HAVING THE
FEELING IT MIGHT BE A PERIOD OF ABOUT AN HOUR OR TWO AT THE END
OF THE DAY WHICH OBVIOUSLY WON‘T BE NOW, WHEN WE COULD BRING
THIS PERSON IN FROM THE BOARD OF PARDONS AND PAROLES.
S00 WITH THAT IN MIND, I THINK MR. STROUP PUT HIM UNDER
SUBPOENA FOR TWO O'CLOCK AND WAS IN FACT QUT IN THE HALL HOPING
TO MEET HIM AT TWO O“CLOCK AND SAY YOURE DISMISSED TILL
MONDAY.
THE COURT: GET TO THE FOINT. WHEN DO YOU WANT HIM?
MR. BOGER: WE DON’T WANT HIM TODAY. WE PROBABLY DON’T
NEED HIM UNTIL AFTER THE STATISTICIAN TESTIFIES.
THE COURT: ALL RIGHT. PUT HIM ON CALL AND INDICATE TO
HIM THAT HE WON‘T BE REACHED MONDAY AND HE SHOULD BE AVAILABLE
TUESDAY, WEDNESDAY AND THURSDAY.
MS. BOLEYN: THANK YOU, YOUR HONCR.
MR. BOGER: THANK YOU, YOUR HONOR.
— —————. — ——
977
BALDUS - DIRECT
DAVID C. BALDUS,
BEING PREVIOUSLY DULY SWORN, RESUMED THE WITNESS STAND AND
TESTIFIED FURTHER AS FOLLOWS:
DIRECT EXAMINATION (CONT’D)
BY MR. BOGER:
@. PROFESSOR BALDUS, AS YOU'RE AWARE, THE PETITIONER IN THIS
CASE IS WARREN MCCLESKEY. WHO WAS TRIED, CONVICTED AND SENTENCED
IN FULTON COUNTY HERE IN GEORGIA.
DID YOU AT ANY POINT RECEIVE A REQUEST FROM ME OR
ANYONE ELSE TO CONDUCT ANY ANALYSES DIRECTED ONLY TOWARD FULTON
COUNTY, EMPLOYING THE DATA YOU‘D COLLECTED IN YOUR CHARGING AND
SENTENCING AND PROCEDURAL REFORM STUDIES? |
A. YES, 1 RECEIVED SUCH A REQUEST.
Q. AND DID YOU IN FACT CONDUCT ANY ANALYSIS?
A. YEE, I DID.
2. WHAT WAS THE NATURE OF YOUR ANALYSIS? HOW DID YOU PROCEED?
A. WELL, THE OBJECT OF THE ANALYSIS WAS TO DETERMINE WHETHER
THE SENTENCING PATTERNS THAT WE HAD OBSERVED STATEWIDE WITH
RESPECT TO RACE OF VICTIM AND RACE OF DEFENDANT DISPARITIES ALSO
[PERTAINED IN FULTON COUNTY. AND WHETHER THE EVIDENCE CONCERNING
FULTON COUNTY SHED ANY LIGHT ON THE QUESTION OF WHETHER WARREN
MCCLESKEY-S DEATH SENTENCE WAS AN ABERRANT DEATH SENTENCE. OR
WHETHER RACIAL CONSIDERATIONS MAY HAVE PLAYED A ROLE IN THE
DISPOSITION OF HIS CASE.
— —— — ———" —— —
78
BALDUS - DIRECT
1 RQ. HOW DID YOU BEGIN YOUR ANALYSIS?
2 ki WE BEGAN THE ANALYSIS BY REPRODUCING, OR REPLICATING THE
3 TYPE OF ANALYSIS THAT WE HAD DONE STATEWIDE WITHIN THE
4 CONFINES OF FULTON COUNTY AND THAT WAS DONE IN THE. WITH,
bo’ RATHER, THE DATA FROM THE CHARGING AND SENTENCING STUDY.
) WE SIMPLY IDENTIFY THE CASES IN WHICH SENTENCE WAS
7 IMPOSED IN THIS COUNTY AND PROCEED WITH THE ANALYSIS OF THOSE
8 CASES.
4 QA. LET ME DIRECT YDUR ATTENTION NOW TO DB-106 MARKED FOR
0 IDENTIFICATION. ILL ASK YOU IF YOU CAN IDENTIFY THAT DOCUMENT?
11 A. YES. THIS IS A BREAKDOWN BY STAGE IN THE PROCESS OF THE
12 CASES IN FULTON COUNTY. AND IT, IT INDICATES THE RATES AT WHICH
13 CASES ARE DISPOSED OF AT EACH STEP IN THE PROCESS. AND FINALLY
14 LOOKING AT THE FOOT OF COLUMN B, IT INDICATES THE OVERALL DEATH
15 SENTENCING RATE IN THE COUNTY.
16 FURTHERMORE, IT PRESENTS THESE RATES AT EACH STAGE IN
17 THE PROCESS, BROKEN DOWN BY THE DEFENDANT-VICTIM RACIAL
18 COMBINATION.
w 1% RN. WHAT DOES DB-1046 REFLECT?
20 A. DB-104 SHOWS THAT WHEN YOU LOOK AT THE COMBINED EFFECTS OF
23 CECISIONS AFTER MURDER INDICTMENTS RESULTING IN A DEATH
<2 SENTENCE, THAT THE LIKELIHOOD OF A DEATH SENTENCE IS HIGHER IN
23 WHITE VICTIM THAN BLACK VICTIM CASES.
24 IT ALSO SUGGESTS THAT AT SOME OF THE EARLIER STAGES IN
25 THE PROCESS, THAT DEFENDANTS WITH WHITE VICTIMS HAVE A HIGHER
erer——— ——— ———————— —
RTP
BALDUS - DIRECT
1 CHANCE OF BEING RETAINED IN THE SYSTEM AND CONTINUING TO RUN A
RISK OF CAPITAL PUNISHMENT.
Q. DOES DB-10& THEN SUGGEST ANY HYPOTHESIS ABOUT RACIAL IMPACTS
IN FULTON COUNTY?
tla YES. IT DOES SUGGEST A HYPOTHESIS THAT RACIAL FACTORS,
SPECIFICALLY THE RACE OF THE VICTIM, MAY PLAY A ROLE IN THE
DECISION MAKING PROCESS HERE.
8 IT DOES NOT. HOWEVER. TEND TO SUPPORT ANY SUGGESTION
S |THAT THE RACE OF THE DEFENDANT DOES.
0 |@. DOES THIS TABLE TO DB-104 REFLECT UNADJUSTED NUMBERS OR HAVE
11 |YOU ADJUSTED FOR ANYTHING THERE, BEYOND THE RACIAL
12 |VICTIM-DEFENDANT COMBINATIONS?
13 |A. NO. THESE ARE STRICTLY UNADJUSTED DATA PROVIDED FOR PURPOSES
14 |OF GIVING ONE AN OVERVIEW OF THE SYSTEM AND SUGGESTING
1S |HYPOTHESES ABOUT THE POTENTIAL ROLE OF RACIAL FACTORS IN THE
14 |svsvem.
17 '@. WAS THIS YOUR FIRST STEP IN THE ANALYSIS OF FULTON COUNTY?
12 lA. YES. IT WAS.
» 19 MR. BOGER: YOUR HONOR, 1 MOVE THE ADMISSION OF DB-106
20 INTO EVIDENCE.
21 | THE COURT: FOR THE PURPOSE OF DEMONSTRATING THE
22 PROCESS OF ANALYSIS, IT WILL BE ADMITTED, NOT FOR THE TRUTH OF
23 THE MATTER ASSERTED THEREIN.
24 BY MR. BOGER?
2% A. NOW. PROFESSOR BALDUS, DID YOU THEN GO ANY STEP FURTHER AND
— — p——_ a—————— —— —— ——— ——— —— — ——
280
BALDUS - DIRECT
1 SHARPEN OR FOCUS YOUR ANALYSIS TO BEGIN TO TEST THE HYPOTHESIS
BJ
SUGGESTED BY DE-1067
a Ia. YES. 1 DID.
3 @. HOW DID YOU DO THAT?
(&
A. 1 FOCUSED ON THE HYPOTHESIS SUGGESTED BY THE PRECEDING
ANALYSIS, AND THE RESULT OF THAT ANALYSIS IS PRESENTED IN
{Q. WHAT DOES DB-107 SHOW?
y
o
S
Y
BR
E
1
A. DB-~107 SHOWS UNADJUSTED DISPARITIES IN THE DISPOSITION OF
10 CASES AT SIX POINTS IN THE CHARGING AND SENTENCING PROCESS IN
11 FULTON COUNTY.
12 IT SHOWS THAT AT EACH STAGE IN THIS PROCESS THAT THERE
13 IS A DIFFERENTIAL TREATMENT IN THE DISPOSITION OF WHITE AND
14 BLACK VICTIM CASES.
13 I CAN ELABORATE ON THAT. AT GUILTY PLEA STAGE. BLACK
146 VICTIM CASES HAVE A HIGHER LIKELIHOOD OF RESULTING IN A
17 VOLUNTARY MANSLAUGHTER PLEA FOLLOWING A MURDER INDICTMENT.
i2 @. WHAT’S THE DIFFERENCE THERE?
w 19 A. ITS MINETEEN PERCENTAGE POINTS DIFFERENCE.
20 AND WITH RESPECT TO GUILTY FLEAS TO MURDER, THE RATE.
21 THE WHITE VICTIM CASES RESULT IN PLEAS TO MURDER AT A HIGHER
22 RATE THAN DO BLACK VICTIM CASES, WHICH IS JUST THE OFFOSITE WITH
23 RESPECT TO THE PLEAS TO VOLUNTARY MANSLAUGHTER.
24 AT THE GUILT TRIAL. THE CONVICTION RATE IS HIGHER FOR
23 WHITE VICTIM CASES FOR MURDER. OUR DATA, AS THEY RELATE TO THIS
BALDUS = DIRECT
1 DECISION POINT. ONLY ALLOW FOR AN ANALYSIS OF TWO DECISIONS.
2 THAT 1S. WHETHER THERE WAS A MURDER CONVICTION OR A VOLUNTARY
3 MANSLAUGHTER CONVICTION. AND THERE IS A NINETEEN PERCENTAGE
4 POINTS HIGHER RISK OF A MURDER CONVICTION IN A WHITE VICTIM
> CASE.
8 & IN TERMS OF THE ADVANCEMENT OF THE CASES TO PENALTY
¥ TRIAL, THE. RATE IS SOME ELEVEN POINTS HIGHER IN WHITE VICTIM
8 CASES.
4 AND IN TERMS OF DEATH SENTENCING DECISIONS IT’S HIGHER
10 BY SOME 48 PERCENTAGE POINTS.
11 THE OVERALL EFFECT IN THE SYSTEM IS APPROXIMATELY FIVE
12 PERCENTAGE POINTS, AND THAT DISPARITY IS STATISTICALLY
13 SIGNIFICANT AT THE .0S5 LEVEL. THAT IS. THE EFFECTS OF THE, THE
14 DISPARITIES CALCULATED IN TERMS OF THE COMBINED EFFECTS OF ALL
1S |OF THESE DECISION POINTS.
16 AGAIN 1 ADD, HOWEVER, THAT THESE ARE DISPARITIES
17 CALCULATED WITHOUT ADJUSTMENT FOR ANY BACKGROUND FACTORS.
18 MR. BOGER: YOUR HONOR, AT THIS TIME, I MOVE THE
Ww 319 ADMISSION OF DB-107.
20 MS. WESTMORELAND: YOUR HONOR, I HAVE THE SAME
21 (OBJECTIONS. THIS APPEARS AGAIN TO BE, I PRESUME A FOUNDATION
«2 DOCUMENT AS THE ONE BEFORE IT, SIMILAR TO THE ONE BEFORE IT.
23 | THE COURT: YES. TO DEMONSTRATE PROCESS. IT WILL EE
24 ADMITTED, NOT FOR THE PROOF OF THE MATTERS ASSERTED THEREIN.
25 MR. BOGER: YOUR HONOR, I SUPPOSE 1 SHOULD RENEW OUR
«a
ag
.
0
M
0
Si —— | — —— —— —— —— —— —— r——— —————
82
BALDUS - DIRECT
CONTENTION WHICH WE MADE EARLIER AND WHICH I“M AWARE YOUR HONOR
OVERRULED, THAT ITEMS SUCH AS THIS, WHICH DON’T CONTROL FOR A
SIGNIFICANT NUMBER OF BACKGROUND FACTORS ARE NOT ADMITTED FOR
ALL PURPOSES, AT LEAST THAT’S WHAT I UNDERSTAND TQ BE YOUR
HONORS PREVIOUS RULINGS.
WE NOW HAVE FOCUSED DOWN FROM THE STATEWIDE SYSTEM ON
TO FULTON COUNTY. AND IT“S MY SUBMISSION TO THE COURT THAT EVEN
IF YOUR HONDR’S RULING IS CORRECT WITH RESPECT TO THE STATEWIDE
RULING, AND. OF COURSE, WE HAVE DISAGREED ON THAT» THAT WHEN ONE
REACHES A COUNTY LEVEL SUCH AS THIS. EVEN UNADJUSTED DATA OF
THIS SORT IS RELEVANT AND SHOULD BE ADMITTED FOR THE TRUTH
THEREDF, AS WELL AS TO SHOW THE PROCESS THAT FROFESSOR BALDUS
HAS FOLLOWED.
I THINK WE’RE GOING TQ FIND THAT BECAUSE OF SMALLER
NUMBER OF CASES. THAT THE DIFFERENT RULES PERHAPS SHOULD APPLY
AT THIS STAGE THAN APPLY AT A STATEWIDE STAGE WHERE THERE’S A
LARGER BASE TO DRAW FROM. THAT S MY CONTENTION.
THE COURT: I UNDERSTAND IT. I HAVE NOT ADMITTED IT.
MR. BOGER: ALL RIGHT, YOUR HONOR, FINE. WE“LL
CERTAINLY ABIDE BY THAT.
BY MR. BOGER:
QR. NOW, PROFESSOR BALDUS, DID YOU INTRODUCE ANY BACKGROUND
CONTROLS IN LOOKING AT THESE RACE OF VICTIM DISPARITIES IN
FULTON COUNTY?
A. YES. WE STARTED THE PROCESS BY CONTROLLING FOR ONE
283
| BALDUS ~- DIRECT
P
Y
| STATUTORY AGGRAVATING FACTOR AT A TIME. AND THE RESULTS OF
THOSE ANALYSES ARE PRESENTED IN DB-108, WHICH IS TABLE 4 FROM
THE FULTON COUNTY REPORT.
E
E
'Q. WHAT DO THOSE RESULTS SHOW?
A. THNSE RESULTS SHOW A CLEAR PATTERN OF RACE OF VICTIM a
& DISPARITIES IN DEATH SENTENCING RATES AMONG THE CASES WHICH OUR
7 ANALYSIS SUGGESTED WERE DEATH ELIGIBLE UNDER EACH OF THESE
3 INDIVIDUAL FACTORS. AND THE NUMBERS, THE NUMBER OF THE
° CATEGORIES ARE NOT VERY LARGE, BUT WE‘RE DEALING WITH A VERY
0 SMALL NUMBER OF DEATH SENTENCES, ONLY TEN DEATH SENTENCES IN THE
11 SAMPLE, SO THAT THE IMPORTANT THING HERE IS TO EXAMINE THE
12 PATTERN, AND THE PATTERN, I THINK, IS PLAIN, THAT THERE ARE
13 DISPARITIES IN THE DISPOSITION OF THESE CASES THAT TEND TO TREAT
14 THE, TREAT THE WHITE VICTIM CASES MORE PUNITIVELY THAN THE BLACK
135 VICTIM CASES.
146 A. LET ME DIRECT YOUR ATTENTION PARTICULARLY. PROFESSOR BALDUS,
12 TQ THE B2 AGGRAVATING CIRCUMSTANCE AND THE BS AGGRAVATING
18 CIRCUMSTANCE.
ye 17 CAN YOU INDICATE THE RESULTS IN THOSE TWD AREAS?
20 A. YES. THE ESTIMATED EFFECTS IN TERMS OF THE B2 ANALYSIS IS
21 NINE PERCENTAGE POINTS DIFFERENCE. AND THE ESTIMATED DISPARITY
lL | IN THE BS CATEGORY IS TWENTY PERCENTAGE POINTS.
23 | BUT PARTICULARLY IN THIS LATTER CATEGORY » I WANT TO
24 [EMPHASIZE THAT THOSE ARE VERY SMALL NUMBERS. HOWEVER, IN THE BZ
23 CATEGORY, THE ESTIMATED NUMBERS ARE LARGER, ALTHOUGH THERE, THE
<0
¥84
BALDU3 - DIRECT
DISPARITY BETWEEN THE TWO RATES IS NOT, IT’S SIMPLY NINE POINTS.
@. LET ME LOOK AS WELL AT B10, KILLING TO AVOID ARREST, WHICH
yOu HAD INDICATED IN YOUR PREVIOUS TESTIMONY WAS CORRELATED WITH
THE B8 FACTOR?
A. YES. WHAT WE SEE THERE IS A NINETEEN PERCENTAGE POINT
DIFFERENCE IN THE DEATH SENTENCING RATES.
@. AND IN ALL THESE THREE THAT WE-’VE LOCKED AT, B2, E28 AND B10,
‘WHICH CATEGORY OF OFFENDER RUNS THE GREATEST RISK OF A DEATH
SENTENCE AS REFLECTED IN THESE FIGURES?
A. OH, IN THESE FIGURES IT SUGGESTS THAT THE WHITE VICTIM CASES
"RUN A HIGHER RISK OF RECEIVING THE DEATH SENTENCE.
MR. BOGER: YOUR HONOR. 1 MOVE THE ADMISSION OF DB-108
INTD EVIDENCE. |
MS. WESTMORELAND: SAME OBJECTION AS NOTED TO THE LAST
TWO, YOUR HONOR.
THE COURT: SUSTAIN THE OBJECTION.
RATHER THAN PLAYING CAT AND MOUSE WITH YOU, UNTIL YOU
GET INTO BALANCING THE SUBSTANTIAL FACTORS AS DEMONSTRATED
EITHER BY A KNOWLEDGE OF THE CRIMINAL JUSTICE SYSTEM OR BY ANY
OTHER RELIABLE STATISTICAL TECHNIQUE, ONE OF WHICH MAY BE THE
STEPWISE, I DON’T THINK WE‘RE DEALING IN MUCH OF PROBATIVE
VALUE.
HAVE YOU GOT SUCH?
MR. BOGER: YOUR HONOR, I THINK WE HAVE THINGS THAT
WILL ADVANCE US A GREAT DEAL FURTHER ALONG IN THAT DIRECTION.
| 983
BALIUS - DIRECT
1 IM NOT TRYING TO TO PLAY CAT AND MOUSE EITHER. 1 DQ
2 WANT THESE EXHIBITS INTO EVIDENCE AND I UNDERSTAND YOUR HONORS
3 RULING AND JUST FEEL A NEED TD MAKE THE RECORD ON THAT AS WELL.
4 AND DO THINK THESE DO ADVANCE THE ANALYSIS.
ba THE COURT: HE CAN MAKE HIS OPINION BASED ON ANYTHING
5 HE WANTS TO, BUT I“M NOT GOING TO LET THEM IN EVIDENCE IF
7 | THEYRE NOT MINIMALLY ADJUSTED TO FACTORS WHICH EVERYBODY KNOWS
8 COME INTO PLAY BEYOND THE FEW FACILE THINGS THAT WE“VE DONE SO
? FAR.
10 MR. BOGER: YOUR HONOR, LET ME JUST CLARIFY YOUR RULING
i1 ON 108,
12 ON DB-10& AND DB~107 I BELIEVE YOU INDICATED THEY WERE
13 ADMITTED TO SHOW THE PROCESS BUT NOT FOR THE TRUTH THEREFORE.
14 IS 108 ALTOGETHER EXCLUDED OR IS IT ADMITTED ON THAT LIMITED
13 BASIS AS WELL?
146 THE COURT: I WOULD ASSUME THAT IT IS PART OF THE
A7 HYPOTHESIS BUILDING FROCESS BECAUSE 1M SEEING THE SAME
13 RECURRENCE OF STEPS THAT I BELIEVE I SAW EARLIER. AND IF YOU & 19 |WANT IT FOR THAT REASON, I“LL GIVE IT TO YOU FOR THAT REASON.
20." MR. BOGER: THANK YOU YOUR HONOR.
|
21 |BY MR. BOGER:
22 RQ. NOW FROFESSOR BALDUS, DID YOU TRY TO CONTROL FOR ANY OTHER
23 BACKGROUND VARIABLES BEYOND THESE INDIVIDUAL STATUTORY
24 AGGRAVATING CIRCUMSTANCES CONSIDERED ONE AT A TIME?
23 8. YES. ‘1 LIN.
86
BALDUS - DIRECT
@. AND HOW DID YOU DO THAT?
A. WELL, ONE METHOD WAS TO IDENTIFY GROUPS OF SIMILAR CASES,
[dross OF CASES THAT OUR ANALYSIS SUGGESTED HAD A COMPARABLE
LEVEL OF OVERALL CULPABILITY AS MCCLESKEY‘S CASE. AND THE WAY
WE COMMENCED THIS PROCESS OF TRYING TO PULL TOGETHER GROUPS OF
COMPARABLE CASES WAS FIRST TO SORT THE CASES IN TERMS OF THE
FACTORS THAT STATEWIDE TENDED TO PREDICT WHO RECEIVED A DEATH
| SENTENCE.
AS I INDICATED EARLIER, THESE REGRESSION ANALYSES THAT
SORTED THE CASES ACCORDING TO THEIR PREDICTED LIKELIHOOD OF A
DEATH SENTENCE GIVE ONE A ROUGH MEASURE OF THE AGGRAVATION LEVEL
OF THE CASES THAT ONE IS DEALING WITH, AT LEAST AS MANIFESTED BY
THE DECISIONS OF THE PROSECUTORS AND JURIES WHO HANDLE THESE
CASES.
SO, WE PRODUCED A SORT OF THE CASES FROM FULTON COUNTY
ON THE BASIS OF THESE INDICES. NOT PLURAL, SINGULAR. THE INDEX
THAT PREDICTED THE LIKELIHOOD OF A DEATH SENTENCE GIVEN A MURDER
INDICTMENT.
Q. HOW WAS THIS INDEX CONSTRUCTED?
A. THE INDEX IS CONSTRUCTED BY EXAMINING THE STATEWIDE ANALYSIS
WHICH PREDICTS WHO RECEIVES A& DEATH SENTENCE.
THEN EXAMINING THE INDIVIDUAL FACTS OF EACH CASE IN
FULTON COUNTY. AND ASSIGNING IT A SCORE WHICH MEASURES THE
LIKELIHOOD THAT THAT CASE WOULD RECEIVE THE DEATH SENTENCE,
GIVEN THE FACTS OF THAT CASE.
87
BALDUS - DIRECT
1 | THOSE CASES WERE THEN RANK ORDERED ACCORDING TO THE
2 PREDICTED LIKELIHOOD THAT THOSE OFFENDERS WOULD RECEIVE A DEATH
SENTENCE, AND WE THEN IDENTIFIED THE CASES AMONG THAT GROUP THAT
4 WERE COMPARABLE IN TERMS OF A PREDICTED LIKELIHOOD TO MCCLESKEY.
be @. HOW MANY FACTORS WERE TAKEN INTO ACCOUNT IN DEVISING THIS
& INDEXING SCORE?
yj A. THENTY,
8 RA. WHAT WERE THEY?
2 A. COUNSEL, THESE ARE LISTED AS A FOOTNOTE IN THE FULTON COUNTY
10 REPORT.
11 WOULD YOU LIKE ME TO READ THEM?
12 A. YES, I WOULD, IF YOU DON‘T HAVE A PRESENT RECOLLECTION?
13 A. NO, I DO NOT.
14 WAS THERE A RACE HATRED MOTIVE IN THE CASE.
15 WAS THE DEFENDANT A TRIGGER MAN.
16 DEFENDANT EXPRESSED PLEASURE WITH THE KILLING.
17 VICTIM FORCED TO DISROBE.
13 WHAT WERE THE NUMBER OF STATUTORY AGGRAVATING
Ww 19 |CIRCUMSTANCES IN THE CASE.
20 WAS THE MOTIVE TO AVENGE THE ROLE PLAYED BY A D.A. OR
21 |LAWYER.
Se WAS THERE AN INSURANCE MOTIVE.
23 | WAS THERE A KIDNAPPING INVOLVED.
24] CO-PERPETRATOR RECEIVE A LESSER SENTENCE.
DID THE DEFENDANT ADMIT GUILT AND ASSERT NO DEFENSE. PJ
&]
Fa
BALDUS - DIRECT
I.
WAS THERE AN INSANITY DEFENSE.
DID THE DEFENDANT FANIC IN THE COURSE OF A BURGLARY.
| DID THE DEFENDANT HAVE IMPAIRED MENTAL CAPACITY.
PARDON ME.
WAS THE VICTIM BEDRIDDEN OR HANDICAPPED.
THE NUMBER OF MAJOR TYPES OF INCRIMINATING EVIDENCE
AGAINST THE DEFENDANT.
NUMBER OF MINOR NON-STATUTORY AGGRAVATING FACTORS IN
o
.
o
o
R
d
THE CASE.
10 NUMBER OF VICTIM-RELATED MITIGATING CIRCUMSTANCES.
ii WAS THERE TORTURE IN THE CASE.
12 THE NUMBER OF MAJOR NON-STATUTORY AGGRAVATING FACTORS
13 IN THE CASE.
14 AND THE STATUS OF THE VICTIM.
13 iQ. HOW WERE THOSE PARTICULAR FACTORS SELECTED AGAIN? I WANT TO
16 BE CLEAR ON THEM.
17 A. THEY WERE COLLECTED OR IDENTIFIED IN A MULTIPLE REGRESSION
18 ANALYSIS OF THE CASES IN THE STATE OF GEORGIA. ALL OF THE CASES
& 19 IN THE STATEWIDE SAMPLE.
20 THE COURT: AS WHAT?
21 THE WITNESS: AS FACTORS WHICH PREDICT WHO GETS DEATH
at SENTENCES THROUGHOUT THE STATE OF GEORGIA.
a THE COURT: AS THE TWENTY THE MOST PREDICTIVE OR ——
24 THE WITNESS: YES, YOUR HONOR, THE TWENTY THAT ARE THE
23 MOST PREDICTIVE, WHICH SATISFACTORY. AGAIN, THE CRITERION WE
Fa?
BALDUS - DIRECT
|
LAID DOWN OF BEING NON-SUSPECT. AND BEING AGORAVATING FACTORS
THAT HAVE AN AGGRAVATING EFFECT, AND MITIGATING FACTORS THAT
HAVE A MITIGATING EFFECT.
BY MR. BOGER?:
Q@. AND BY NOT SUSPECT. YOU MEAN WHAT?
A. IT’S NOT THE SEX OF THE DEFENDANT. OR THEIR SOCIO-ECONCMIC
STATUS OF THE DEFENDANT, MATTERS OF THAT SORT. EDUCATIONAL LEVEL
OF THE DEFENDANT.
@. ONCE YOU DEVELOPED THIS INDEX. WHAT DID YOU DO WITH THE
CASES IN FULTON COUNTY?
A. HAVING IDENTIFIED THIS. OH. INDEX. WE APPLIED IT TO THE
FULTON COUNTY CASES. AND AS 1 SUGGESTED EARLIER, THIS PRODUCED A
LISTING OF ALL THE CASES IN FULTON COUNTY, RANK ORDERED IN TERMS
OF HOW LIKELY THOSE FACTORS AND THE SCORES THAT HAD BEEN
GENERATED FOR THOSE FACTORS IN THE STATEWIDE ANALYSIS WERE TO
RECEIVE A DEATH SENTENCE IN THE FULTON COUNTY SAMPLE.
2. WHAT METHOD DID YOU EMPLOY TO APPLY THE INDEX TO THE FULTON
COUNTY CASES? HOW DID YOU DO IT?
‘A. WELL, WE PICKED THE. AFTER HAVING LISTED ALL THE CASES IN
FULTON COUNTY. WE IDENTIFIED WARREN MCCLESKEY. AND THEN
IDENTIFIED THE CASES, 27 CASES THAT WERE CLOSEST TO WARREN
MCCLESKEY IN TERMS OF THIS PREDICTED LIKELIHOOD OF A DEATH
SENTENCE.
WE DEFINED THESE AS THE NEAR NEIGHBORS TO MR. MCCLESKEY
IN TERMS OF THIS LISTING.
©
~N
P
O
3
10
20
¥90
BALDUS - DIRECT
@. AND WHAT’S THE RATIONALE FOR DEFINING NEAR NEIGHBORS?
A. THE NOTION HERE IS TO TRY AND FIND OFFENDERS WITH COMPARABLE
LEVEL OF CULPABILITY IN A CASE. AND THIS IS THE SAME METHOD THAT
WE USED EARLIER IN OUR ANALYSES, TO TRY AND SORT TOGETHER
CATEGORIES OF CASES WITH SIMILAR LEVELS OF CULPABILITY.
AS I INDICATED ONE WAY IS TO MATCH CASES STRICTLY IN
TERMS DF THEIR FACTUAL COMPARABILITY, BUT IN OUR DATA FILE, WE
ONLY HAD, THAT IS, THE DATA FILE FROM THE CHARGING AND
SENTENCING STUDY. WE ONLY HAD A COUPLE OF CASES INVOLVING POLICE
OFFICERS. S0 WE COULDN‘T FOCUS STRICTLY IN THAT FILE ON CASES IN
WHICH THE VICTIM WAS A POLICE OFFICER.
WE HAD TO EXPAND OUR NET, IF YOU WILL. TO TRY AND FIND
OTHER CASES WHICH HAD AN OVERALL CULPABILITY LEVEL THAT APPEARED
COMPARABLE TO MCCLESKEY, AND THATS ONE OF THE FUNCTIONS THAT
THAT INDEX PROCEDURE SERVES. IT ENABLES YOU TO WEIGH THE
AGGRAVATING AND THE MITIGATING FACTORS OF AN INDIVIDUAL CASE
AGAINST ONE ANOTHER AND PRODUCE AN OVERALL APPROXIMATION OF WHAT
THE CULPABILITY LEVEL OF THE CASE IS.
@. ONCE YOU HAD DONE THAT. WHAT THEN DID YOU DO IN YOUR
ANALYSIS?
A. WELL, WE FOUND THAT USING THIS METHOD THAT HAVING FICKED THE
27 CASES THAT WERE COMPARABLE TO MCCLESKEY., THAT WE FOUND THAT
THERE WERE ANOTHER FIVE PENALTY TRIAL CASES WHICH DID NOT MAKE
THEIR WAY ON TO THIS LIST.
BR. THEIR INDEX SCORE WAS NOT HIGH ENOUGH TO GET ON THE LIST?
BALDUS ~ DIRECT
1 A. THAT’S RIGHT, THEIR INDEX SCORE WAS NOT HIGH ENOUGH TO GET
2 ON THE LIST, BUT NEVERTHELESS THEY HAD A PENALTY TRIAL.
3 AND SINCE OUR OBJECT WAS TO TRY AND IDENTIFY PEOPLE WHO
4 HAD RUN A COMPARABLE RISK OF RECEIVING A DEATH SENTENCE TO
3 MCCLESKEY. WE FIGURED. WE DETERMINED THAT IT WAS APPROPRIATE TO
4 6 INCLUDE THOSE CASES IN THIS CATEGORY OF NEAR NEIGHBORS IN TERMS
z OF COMPARABILITY ON THE ISSUE OF RISK OF A DEATH SENTENCE. SO
38 THAT LEFT US WITH A POOL OF 32 CASES, INCLUDING MCCLESKEY.
? BG. WHAT DID YOU DO WITH THE POOL OF CASES?
10 A. WE THEN READ THE SUMMARIES. AS I MENTIONED TO YOU EARLIER.
11 ONE OF OUR PRINCIPAL APPROACHES IN HANDLING THESE CASES WAS TO
12 LOOK AT THE SUMMARY OF EACH CASE.
13 WE LOOKED AT THE SUMMARIES OF ALL THESE OFFENDERS AND
14 DEVELOPED A SYSTEM OF CATEGORIZATION, TO SORT THEM INTO LEVELS
15 WHICH WE CONSIDERED MORE AND LESS AGGRAVATED THAN MCCLESKEY.
16 @. HOW WERE THOSE DETERMINATIONS OR BY WHOM WERE THOSE
17 DETERMINATIONS MADE?
13 A. THESE DETERMINATIONS WERE DONE BY ME, AND THE, THE WORK OF
Ww 19 SORTING OUT THE CASES WAS DONE BY A RESEARCH ASSISTANT WORKING 20 WITH ME. BUT THE FINAL DETERMINATION OF WHAT CRITERIA WE SHOULD
21 USE TO DETERMINE MORE OR LESS AGGRAVATED CASES WAS MADE BY ME.
|
22 | THE COURT: LET ME ASK YOU A QUESTION. YOU SAID AND TO
23 IDENTIFY THEM AS BEING MORE OR LESS AGGRAVATED THAN MCCLESKEY?
24 THE WITNESS: YES, YOUR HONOR.
ka
a THE COURT: OKAY, 1 SEE. ALL RIGHT. I UNDERSTAND. I
, ——— 1 Av——— —— ——— " : os r—— fp Si——_Sp———— re ——— tet T—— —— = | r—— — —— —— ——— —
e
A
BALDUS ~ DIRECT
f
y
WAS LOOKING AT THOSE GOOSE EGGS ACROSS THE BOTTOM. AND WONDERING
IF YOUU HAD IN FACT CLASSIFIED ANY AS BEING LESS AGCORAVATED. BUT
THEN 1 SEE YOU HAVE PUT ALL THOSE WHERE THE DEATH PENALTY WAS
NOT IMPOSED IN THAT CATEGORY.
BY MR. BOGER:
R. LET ME DIRECT NOW YOUR ATTENTION, FROFESSOR BALDUS. TO
DB-109 AND ASK YOU IF YOU CAN EXPLAIN WHAT THIS FIGURE
REPRESENTS. OR TABLE REPRESENTS?
v
OO
~N
O0
0
A
L
N
A. DB-109, WHICH IS TABLE 5 IN THE FULTON COUNTY REPORT,
fo
o o FRESENTS THE UNADJUSTED AND ADJUSTED RACE OF VICTIM DEATH
Py
P
N
SENTENCING RATES AMONG THE CASES IDENTIFIED AS NEAR NEIGHBORS IN
i2 THE ANALYSIS THAT I HAVE JUST DESCRIBED.
13 | THE FIRST ROW SHOWS AMONG THIS CATEGORY OF CASES THAT
14 THERE WAS A DEATH SENTENCING RATE OVERALL OF .32.
15 AND THAT THE RATE OF THE WHITE VICTIM CASES WAS .47 AS
16 OPPOSED TO .13 FOR THE BLACK VICTIM CASES.
17 AND THAT PRODUCED A DIFFERENCE IN DEATH SENTENCING
13 RATES OF .34.
@ 1? NOW THOSE CASES WERE SELECTED ON THE BASIS OF THE
20 BACKGROUND CONTROLS THAT I JUST DESCRIBED, WHICH WAS EMBODIED IN
21 THE INDEX CREATED FROM THE VARIABLES, WHOSE NAMES I HAVE JUST
<2 READ.
23 AS 1 INDICATED, HOWEVER. WE CONSIDERED IT AFPROFPRIATE
24 TO DO A FURTHER SUBDIVISION OF THESE CASES, WHICH WE DID, AND
2% PARTS 2A THROUGH -C OF TABLE 5 PRESENT THESE RESULTS.
WT
H
e
d
+
Sh
+
BE
R
E
E
8
eg
10
93
BALDUS ~ DIRECT
I SHOULD NOTE THAT THIS SYSTEM OF CATEGORIZATION WAS
GENERATED AS WAS THE SYSTEM OF CATEGORIZATION USED IN THE
ANALYSIS OF THE B2 CASES WITHOUT REGARD TO RACIAL
CHARACTERISTICS OF THE CASES OR THE SENTENCING OUTCOME OF THE
CASE.
THE RESULTS SHOW THAT AMONG THE MORE AGGRAVATED CASES
THAN MOCCLESKEY. 1 MENTION PARENTHETICALLY THAT MCCLESKEY IS IN
CATEGORY 2B. BEING CHARACTERIZED AS THE TYPICAL CASE, THAT THE
MORE AGGRAVATED CASES, INCLUDED FIFTEEN OFFENDERS, AND THAT
THERE WAS A FORTY-TWO PERCENTAGE POINTS DISPARITY IN THE RATES
AT WHICH BLACK AND WHITE OFFENDERS WERE SENTENCED TO DEATH IN
THAT CATEGORY.
IN MCCLESKEY“S CATEGORY, 2B. WE SEE THAT THERES A MUCH
LOWER DEATH SENTENCING RATE, AND THAT THE, MCCLESKEY“S GROUF.
THERE WERE, INCLUDING MCCLESKEY, THERE WERE TWO QUT OF FIVE WHO
WERE SENTENCED TO DEATH IN THE WHITE VICTIM CATEGORY, WHERE NONE
OF THE BLACK VICTIM CASES IN THAT RECEIVED THE DEATH SENTENCING
RATE.
AMONG THE LESS AGGRAVATED CATEGORIES. WHICH CONSISTED
OF TEN CASES, THERE WERE NO DEATH SENTENCES IMPOSED.
|e. “BEFORE 1 GO ON TO YOUR CONCLUSIONS AS A RESULT OF
CONSTRUCTING AND ANALYZING THIS TABLE. ID LIKE TO ASK YOU
WHETHER YOU INCORPORATED YOUR CHOICES OF MORE AND LESS
AGGRAVATED CASES FROM THE DESCRIPTIONS, THE SUMMARIES THAT YOU
MARE INTO ANY DOCUMENT?
74
BALDUS - DIRECT
1 Ph. YES. I RIN.
+ MR. BOGER: I’D LIKE TO AS THE COURT REPORTER TO MARK
|
3 FOR IDENTIFICATION AS DB-110, AN EXHIBIT.
4 THE COURT: IS THAT YOUR TERMINOLOGY. NEAR NEIGHBORS?
5 THE WITNESS: NO. ITS SOCIAL SCIENCE TERMINOLOGY, YOUR
6 HONOR. :
7 THE COURT: THANK YOU.
gs MR. BOGER: IM NOT SURE ANYBODY WANTS THIS GROUP AS
9 |NEAR NEIGHBORS. YOUR HONOR. |
10 |BY MR. BOGER:
$1 @. CAN YOU IDENTIFY THAT DOCUMENT. PROFESSOR BALDUS?
12 A. D[OB-110 IS A DOCUMENT WHICH CONTAINS THE SUMMARIES OF THE
13 CASES AND WORKING PAPERS THAT WERE USED TQ CREATE TABLE 5.
14 IT ALSO INCLUDES INFORMATION ON OTHER FULTON COUNTY
13 CASES.
14 @. NOW, I“M GOING TO ASK YOU TO TALK ABOUT THE FIRST PAGE DF
17 THAT DOCUMENT AND THE RATIONALE AND PROCEDURE FOR SORTING THAT
ia IT REFLECTS.
ww 19 MR. BOGER: I WANT TO COURT TO KNOW IVE GOT A COPY OF
20 THAT FIRST PAGE, I DIDNT HAVE A COPY MADE OF THE ENTIRE BOOK.
21 IT DID SEEM LIKE IT WAS ADDING A LOT OF PAPER. RUT IF THE COURT
22 WANTS TO FOLLOW ALONG FROM THIS FIRST PAGE. WHICH IS THE FIRST
23 PAGE OF DB-110.
24 BY MR. BOGER:
_— R. I WANT IT TO BE CLEAR ON THE RECORD, PROFESSOR BALDUS. JUST
| 995
| BALDUS ~ DIRECT
EXACTLY HOW YOU DID THE SORTING, WHAT CRITERIA YOU FOLLOWED FOR
MAKING CASES MORE OR LESS AGGRAVATED?
A. YES. I‘LL RESTATE IT VERY BRIEFLY.
THE OBJECT WAS TO PRODUCE SUB-GROUPS OF CASES THAT WERE
COMPARABLE AND TO ESTIMATE RACIAL EFFECTS WITH RESPECT TO THE
LIKELIHOOD OF A DEATH SENTENCE GIVEN A MURDER INDICTMENT WITHIN
‘THOSE SUB-GROUPS OF COMPARABLE CASES.
THE FIRST SORT WAS BASED UPON THE INDEX THAT REFLECTED
THE OVERALL LIKELIHOOD OF THE DEATH SENTENCE IN THE STATE OF
GEORGIA.
WE. THIS SHOWS THE CASES THAT WERE COMPARABLE TO WARREN
MCCLESKEY IN THAT REGARD.
WE FURTHER IDENTIFIED A GROUP OF FIVE PENALTY TRIAL
CASES THAT DID NOT APPEAR ON THAT LIST.
| WE TOOK THAT GROUP OF 32 CASES, CLOSELY READ THE
cionees, AND ON THE BASIS OF THAT READING, MADE A
DETERMINATION THAT FURTHER CATEGORIZATION OF THOSE CASES COULD
BE MADE IN TERMS OF THE FACTS OF THE CASES AS THEY RELATE TO
OVERALL LEVEL OF CULPABILITY OF THE CASES.
Q@. NOW. ON WHAT BASIS DID YOU PICK THE CRITERIA WHICH REFLECT
WHETHER A CASE WAS MORE AGBRAVATED?
A. IF A CASE WAS CLASSIFIED AZ MORE AGGRAVATED, IF IT INVOLVED
UNNECESSARY PAIN, TORTUE OR TORMENT TO THE VICTIM.
| OR IF IT INVOLVED A HIRED AGENT. RAPE, MULTIPLE
VICTIMS, OR HELPLESS VICTIM.
ie
DEFENDANT WHO WAS NOT THE TRIGGER MAN OR AN ACCIDENTAL KILLING.
3 Fob
BALDUS ~ DIRECT
A CASE WAS CLASSIFIED AS LESS AGGRAVATED IF IT INVOLVED
A DEFENDANT AND VICTIM WITHIN A ESTABLISHED RELATIONSHIP CR A
@. ALL RIGHT. AND ON WHAT BASIS, HOW DID YOU COME UPON THESE
FACTORS AS SORTING THE AGGRAVATING AND MITIGATING -— THE LESS --
THE MORE AGGRAVATED AND LESS AGGRAVATED CASES?
A. I MADE THE JUDGMENT ON THE BASIS OF MY KNOWLEDGE OF THE LAW.
ITS PERCEPTION OF WHAT MAKES CASES MORE AND LESS AGGRAVATED, AS
WELL AS MY ANALYSIS OF DEATH SENTENCING DECISIONS IN GEORGIA AND
ELSEWHERE.
R. HAVING MADE THAT ANALYSIS, HAVING SEPARATED THE CASES INTO
THESE CATEGORIES, AND HAVING OBSERVED DISPARITIES THAT AROSE.
WHATS YOUR CONCLUSION OR WHAT DO YOU THINK DB-109 REFLECTS?
‘A. DB-10% REFLECTS IN THE MORE AGGRAVATED CATEGORY, RACE OF,
RACE OF VICTIM EFFECTS. THE BLACK VICTIM CASES HAVE A LOWER
LIKELIHOOD OF RECEIVING A DEATH SENTENCE THAN THE WHITE VICTIM
CASES AND THE SAME THING HOLDS IN THE TYPICAL CATEGORY. 2B. THAT
A NEXT LESS AGGRAVATED CATEGORY. AND THERE IS, HOWEVER, NO
DISPARITY IN THE LEAST AGGRAVATED CATEGORY.
ALSO THIS PATTERN THAT WE SEE HERE WHERE THERE’S NO
DISPARITY AMONG THE LEAST AGGRAVATED CATEGORY. BUT SOME
DISPARITY IN. AMONG THE MORE AGGRAVATED CASES IS CONSISTENT WITH
THE PATTERN THAT WE HAVE OBSERVED THROUGHOUT THE STATE OF
GECRGIA WITH RESPECT TO THE RACE OF THE VICTIM.
Q. WHAT DOES IT SUGGEST TQ YOU?
|
|
|
146
97
BALDUS - DIRECT
i
‘A. WELL. IT SUGGESTS THE SAME THING THAT THE STATEWIDE DATA
SUGGESTS. THAT THE RACE OF THE VICTIM HAS AN EFFECT IN THE
DECISION MAKING PROCESS.
MR. BOGER: YOUR HONOR, AT THIS TIME, I MOVE THE
ADMISSION OF DB-109 AND DB-110 INTD EVIDENCE.
MS. WESTMORELAND: YOUR HONOR, I OBJECT TO BOTH
DOCUMENTS ON SOME OF THE SAME GROUNDS PREVIOUSLY NOTED, BUT ALSO
ON THE ADDITIONAL CROUND THAT MY CONCERN AS TO THE PARTICULAR
CHARACTERISTICS THAT HAVE BEEN SELECTED APPARENTLY BY PROFESSOR
BALDUS BASED ON HIS OWN JUDGMENT IN THE CASES AS TO WHAT MAY OR
MAY NOT MAKE A CASE MORE AGGRAVATGED OR LESS AGGRAVATED, I’M NOT
SURE THAT THIS TABLE IS RELEVANT IN LIGHT OF THAT EXERCISE OF
JUDGMENT.
THE COURT: HOW HAVE YOU GOT MARCUS WAYNE CHENNAULT
CHARACTERIZED?
THE WITNESS: HE“S IN THE MOST AGGRAVATED CATEGORY,
YOUR HONOR.
THE COURT: HOW ABOUT HOUSE?
THE WITNESS: HE’S UP IN THAT CATEGORY, TOO, YOUR
HONOR.
THE COURT: MCCORQUODALE?
THE WITNESS: YES, SIR. THE MOST AGGRAVATED CATEGORY.
THE COURT: HAVE YOU GOT THE SUMMARIES THERE.
THE WITNESS: YES, SIR.
THE COURT: LET ME LOOK AT THEM. HOW HAVE YOU GOT THEM
BALDUS - DIRECT
Pa
y | ARRANGED?
THE WITNESS: SEE. THEYRE ARRANGED BY THE MOST
| AGGRAVATED FIRST. YOUR HONOR, AND THEN THEY RE ORGANIZED BY
NUMBER.
BUT YOU CAN SEE THEM IN TERMS OF THE LEVEL OF
AGGRAVATION HERE, SUMMARY, AND THEN HERE ARE THE MORE
AGGRAVATED, TYPICAL, AND LESS AGGRAVATED.
MR. BOGER: I MIGHT NOTE FOR THE RECORD, I DIDN‘T
REALIZE UNTIL I SAW PROFESSOR BALDUS PICK IT UP, BUT DB-110 ALSO
FW
Y
Pe
RE
G
B
R
T
Ul
C
e
4
T
B
S
TR
G
GH
LY
CONTAINS SOME ANALYSES OF POLICE OFFICER CASES WHICH WE'RE ALSO
11 GOING TO OFFER. AND, OF COURSE. THAT PORTION OF THE DOCUMENT
12 ‘WERE NOT OFFERING AT THIS TIME.
13 THE COURT: UH HUH.
14 IM NOT GOING TO SIT HERE AND READ ALL oF THESE. I WAS
13 JUST TRYING TO CET A FLAVOR.
16 I THINK THE EASIEST THING TO DO IS TO ALLOW ALL OF THE
17 EVIDENCE OF CARE TO CREATE A SLIGHT PRESUMPTION. AND THAT IS
13 SOME DEGREE OF ACADEMIC HONESTY. TO SAY THAT HAVING EXERCISED
Ww 1? SOME MORE SOPHISTICATED ANALYSIS WITH REFERENCE TO THIS DATA, 1
20 WILL ADMIT IT.
«1 1 HAVE LOOKED AT TWO SUMMARIES. AND CAN ALREADY FIND
22 THAT I WOULD HAVE SOME GUT LEVEL ARGUMENT WITH THE NEAR NEIGHBOR
23 ANALDGY, BUT I THINK THAT MAYBE GOES MORE TO THE WEIGHT THAN IT
24 DOES TO THE ADMISSIBILITY.
25 SO I WILL ADMIT IT.
—— —— — —— — ——— p———— T— An... SO ———... HO—————" ——————r—— —— —— —— ———— a——— p——. S— ———. ———— _ a— B—————
999
BALDUS - DIRECT
IF YOU CAN DEMONSTRATE THAT IT IS NOT AS ADVERTISED, I
WILL BE GLAD TO HEAR FROM YOU ON A MOTION TO STRIKE.
M3. WESTMORELAND: YOUR HONOR, I BELIEVE WE EXAMINED
THIS DOCUMENT IN SYRACUSE, IF I‘M NOT MISTAKEN, BUT I HAVEN’T
SEEN A COPY HERE TODAY.
MIGHT I PERHAPS REQUEST THAT WE COULD BE PROVIDED WITH
AN OPPORTUNITY TO LOOK THROUGH IT SOMETIME DURING THE
PROCEEDING. |
| THE COURT: I THOUGHT I ASKED BOTH OF YOU TO SWAP
DOCUMENTS, ABOUT MONDAY.
MR. BOGER! YOUR HONOR, THIS IS AN ORIGINAL DOCUMENT
THAT WE SIMPLY. I DON’T HAVE A COPY OF IT MYSELF. THEY
CERTAINLY HAD ACCESS TO IT IN SYRACUSE.
THEYVE GOT A REPORT ON THE FULTON COUNTY MATERIAL THAT
MAKES WIDE REFERENCE TO THIS. WE HAVE SUMMARIES OF —-
THE COURT: EXCUSE ME, MR. BOGER. I ASKED YOU ON
MONDAY TO SWAP DOCUMENTS. YOU HAVEN'T. PLEASE DO.
LET/S GO.
MR. BOGER: ALL RIGHT. YOUR HONOR.
BY MR. BOGER?
|
Q. PROFESSOR BALDUS, HAVE YOU CONDUCTED ANY OTHER ANALYSES
| |
BEYOND THOSE IN DB-109 AND DB-110., OR REFLECTED WITH DB-10%,
WHICH LOOK AT RACIAL FACTORS IN FULTON COUNTY CONTROLLING FOR
THE SERIQUSNESS OF THE OFFENSE?
A. YES. I DID A MULTIREGRESSION ANALYSIS. STEPWIZSE REGRESSION
1000
BALDUS - DIRECT
1 ANALYSIS IN FULTON COUNTY ON THE QUESTION OF WHETHER OR NOT
THERE WAS A DEATH PENALTY IMFOSED AMONG CASES FOR MURDER
| INDICTMENT. AND THE EFFECT OF THAT. THE EFFECT OF THAT ANALYSIS
DID NOT SHOW ANY RACE OF VICTIM EFFECT.
QR. CAN YOU EXPLAIN THAT?
A. YES. THE CONSERUENCE OF THE ANALYSIS WAS THAT THERE WAS A
«
>
s
W
O
N
VERY SMALL RACE CO-EFFICIENT ESTIMATED AFTER CONTROLLING FOR
SOME 40 OR 50 VARIABLES, AND IT WAS NOT STATISTICALLY
g
SIGNIFICANT.
10 | THE EXPLANATION FOR THAT. I THINK. IS VERY SIMPLY THAT
i1 IT REFLECTS THE IMPACT ACROSS A WIDE VARIETY OF CASES IN THE
: Bet JURISDICTION, MOST CASES IN THIS JURISDICTION ARE NOT AT ALL
13 AGGRAVATED.
14 I THINK THAT THE POCKETS OF DISPARITY WITH RESPECT TO
13 THE RACE OF THE VICTIM ARE CONCENTRATED STRICTLY AMONG THE MOST
16 AGGRAVATED CASES AND THOSE ARE THE CASES THAT ARE EMBRACED IN
17 THE NEAR NEIGHBOR THAT I JUST PRESENTED.
18 BUT WHEN YOU LOOK AT THE AVERAGE EFFECT OVER THE ENTIRE
, 19 SYSTEM, IT DOES NOT PRODUCE A STATISTICALLY SIGNIFICANT
20 CO-EFFICIENT IN A STEPWISE REGRESSION.
21 @. LET ME ASK YOU TO LDOK AT WHATS MARKED DE-111 FOR
22 IDENTIFICATION.
23 A. UH HUM, YES, DB-1}]1 ——
24 A. EXCUSE ME. DB-11i7
Pte A. 1 BEG PARDON. DB-111 IS A, TABLE NUMBER 3 FROM OUR REPORT.
[8
1001
BALDUS - DIRECT
| AND WHAT IT SHOWS IS AN ANALYSIS CONDUCTED AT EACH LEVEL IN THE
CHARGING AND SENTENCING PROCESS IN FULTON COUNTY. AND THIS IS
AN ANALYSIS THAT ADJUSTS FOR BACKGROUND FACTORS INVOLVED. THAT
EXPLAIN THE DECISIONS AT EACH OF THESE POINTS IN THE PROCESS.
I DESCRIBED HERE EARLIER HOW WE CREATED AN INDEX TO
IDENTIFY COMPARABLE CASES FOR THE CONSTRUCTION OF DB-109.
I REPEATED THAT ANALYSIS WITH RESPECT TO DECISIONS MADE
AT THE EARLIER STAGES IN THE ANALYSIS, PARDON ME, THE EARLIER
STAGES IN THE PROCESS STATEWIDE, AS I CONDUCTED AN ANALYSIS OF
INDICTMENT DECISIONS THROUGHOUT THE STATE. AN ANALYSIS OF PLEA
BARGAIN DECISIONS, VOLUNTARY MANSLAUGHTER. MURDER PLEAS.
DEATH SENTENCING AT JURY TRIAL. LIKELIHOOD THAT A CASE
WOULD BE ADVANCED TQ A FENALTY TRIAL.
AND I THEN APPLIED THE RESULTS OF THOSE ANALYSES TO
RANK ORDER THE CASES IN FULTON COUNTY AT EACH ONE OF THESE
STAGES IN THE PROCESS AND TO IDENTIFY THE CASES THAT WERE
COMPARABLE TO WARREN MCCLESKEY, AT EACH STAGE IN THIS PROCESS.
THE COURT: WHAT ARE YOU TALKING ABOUT, NOW, EXHIBIT
NUMBER -—-— THE WITNESS: 111, YOUR HONOR. DB-11l.
| THE COURT: WELL, WHERE IS YOUR MULTISTEP REGRESSION
ANALYSIS?
| : THE WITNESS: THE RESULTS OF THAT ANALYSIS ARE NOT
\REFORTED HERE. YOUR HONOR. THEY RE NOT STATISTICALLY
SIGNIFICANT, AND THE CO-EFFICIENT WAS VERY SMALL.
a 1002
BALDUS - DIRECT
1 THE ACTUAL TECHNICAL REPORT ON IT, THE SIZE OF THE
2 [CO=EEFICIRNY AND THE LEVEL OF SIGNIFICANCE IS REPORTED IN OUR
3 FINAL REPORT ON FULTON COUNTY.
4 THE COURT: IF THE STATE DOESN-T INTEND TO INTRODUCE
a 17, I WANT IT IN AS PART OF THE RECORD IF WE HAVE TO MARK IT AS
® 4 COURT 1.
7 ALL RIGHT. GO AHEAD.
8 THE WITNESS: THE ——
9 MR. BOGER: LET ME JUST ADD FOR THE RECORD, YOUR HONOR,
10 WE HAD INTENDED, OF COURSE. TO INTRODUCE, AND STILL DO, THAT
11 WHOLE REPORT IF WE CAN AND I SIMPLY WANTED THE RECORD TO BE
12 CLEAR AS WE MOVED ALONG THIS HAD BEEN DONE AND HADN’T HAD THOSE
13 |SIONIFICANT RESULTS.
14 |BY MR. BOGER:
15 !@. PROFESSOR BALDUS, IF YOU —-
16 THE COURT: YOU KEEP SAYING THAT. AND I HAVE
17 | INTENTIONALLY NOT SAID A THING, BECAUSE I DON’T KNOW WHAT YOU“RE
1& | TALKING ABOUT IN TERMS OF A REPORT.
& 19 | MR. BOGER: WE ARE ONE DOCUMENT AWAY FROM HAVING A
20 |DISCUSSION ABOUT THAT MATTER. YOUR HONOR.
21 THE COURT: WELL, LET’S GET ON TO THAT, AND TAKE A
22 |BREAK, AND WE-LL COME BACK AND DISCUSS IT.
23 MR. BOGER: THANK YOU.
24 |BY MR. BOGER:
29 2. PROFFESSOR BALDUS, WHAT DOES DB-111 REFLECT?
14
1003
BALDUS - DIRECT
A. IT REFLECTS THE RESULTS OF THE ANALYSIS, ANALYSES THAT WE
DID AT EACH STEP IN THE PROCESS. TO SUMMARIZE, WE ESTIMATED IN
REGRESSION ANALYSIS. AT BEST PREDICTED. DISPOSITION OF CASES AT
EACH STEP IN THIS PROCESS, AND USING THAT ANALYSIS WE WERE ABLE
TO RANK ORDER THE CASES IN FULTON COUNTY. WITH RESPECT TO THE
LIKELILHOOD OF SELECTION AT EACH POINT IN THIS PROCESS.
ON THE BASIS OF THOSE ANALYSES, WE WERE ABLE TO
IDENTIFY GROUPS OF NEAR NEIGHBORS AT EACH ONE OF THESE SELECTION
POINTS.
AND WE IDENTIFIED WARREN MCCLESKEY-S POSITION IN THE
ORDERING OF CASES AND PULLED OUT THE 2&6 OR 27 CASES THAT WERE
MOST COMPARABLE TO MCCLESKEY.
@. THIS IS THE PROCEDURE YQU’VE ALREADY DESCRIBED?
A. YES, THAT’S RIGHT.
AND THEN WITHIN THOSE CASES. WE WOULD DETERMINE THE
DEATH SENTENCING RATE OR. PARDON ME, NOT DEATH SENTENCING RATE,
BUT AT EACH STAGE WE WOULD IDENTIFY THE RATES AT WHICH
SELECTIONS WERE MADE AMONG BLACK VICTIM AND WHITE VICTIM CASES.
AND THAT'S WHATS PRESENTED IN TABLE 8.
AND AS YOU CAN SEE, THERE ARE, THERE'S A VERY SLIGHT
EFFECT. I MEAN ALL THESE CASES, ALL OF THESE ANALYSES SEES SHOW
RELATIVELY, INVOLVE RELATIVELY SMALL SAMPLES BY VIRTUE OF THE
METHOD WE USED TO TIGHTEN CONTROL FOR COMPARABILITY IN TERMS OF
OTHER FACTORS.
ROW 1 SHOWS A VERY SMALL EFFECT. RACE OF VICTIM EFFECT
1004
BALDUS - DIRECT
i |AT STAGE OF INDICTMENT. NEAR NEIGHBORS.
2 ©. THATS, THE EFFECT IS IN WHAT COLUMN?
3 A. OH. IN COLUMN E., RIGHT.
4 3 O67
5 A. SIX PERCENTAGE FOINTS HIGHER CHANCE OF SELECTION FOR WHITE
& VICTIM CASES.
7 AMONG THE NEAR NEIGHBORS TO WARREN MCCLESKEY., NO ONE
a RECEIVED A PLEA TO VOLUNTARY MANSLAUGHTER OR WAS ALLOWED TO
9 PLEA.
10 THERE WERE AMONG THE NEAR NEIGHBORS TO WARREN
11 MCCLESKEY, THERE WERE TWO PEOPLE WHO PLED GUILTY TO MURDER, AND
12 THOSE WERE. THOSE WERE BLACK VICTIM, RATHER THAN WHITE VICTIM
13 CASES. ALL ARE COMPARABLE DEFENDANTS, WERE CONVICTED OF MURDER,
14 AND THE COMPARABLE CASES WERE, CASES COMPARABLE TO MCCLESKEY
15S |ADVANCED TO PENALTY TRIAL AT ESSENTIALLY THE SAME RATE.
1&4 HOWEVER, WHEN YOU LOOK AT THE DEATH SENTENCES IMPOSED
17 AT PENALTY TRIAL, YOU SEE A DISPARITY OF FORTY-FIVE PERCENTAGE
12 POINTS IN THE RATE THAT DEATH SENTENCES WERE IMPOSED, SEVEN OUT
19 OF TEN OF THE WHITE VICTIM CASES RESULTED IN DEATH SENTENCE,
20 WHEREAS ONLY TWO OUT OF THE EIGHT BLACK VICTIM CASES RESULTED IN
21 A DEATH SENTENCE.
22 A. WHAT DO THESE RESULTS SUGGEST?
23 A. THESE. THESE RESULTS SHOW THAT, AT THESE VARIOUS STAGES,
24 THEY SHOW MILD RACE OF VICTIM EFFECTS. THEY TEND TO EXPLAIN WHY
23 WE SEE AN OVERALL RACE OF VICTIM EFFECT IN TERMS OF DEATH
F
y
ra
ce
BE
E
S
e
B
E
L
Bo
ll
©
eH
©
Ee
2
i
|
|
i
|
100%
BALDUS —- DIRECT
SENTENCING RATES IN FULTON COUNTY FOR CASES INDICTED FOR MURDER.
@. ALL RIGHT.
MR. BOGER: YOUR HONOR, AT THIS POINT I MOVE THE
ADMISSION OF DB-112 -- EXCUSE ME ~- 111 INTO EVIDENCE.
MS. WESTMORELAND: YOUR HONOR, I WOULD LIKE TO ASK ONE
QUESTION CONCERNING ONE OF THE NUMBERS ON TABLE, IF I MIGHT,
NUMBER 4. UNDER COLUMN B. SHOULD THAT BE NINE OUT OF EIGHTEEN
AS OPPOSED TO NINE OUT OF EIGHT, OR AM I READING THE TABLE
WRONG? |
BY MR. BOGERS
G. PROFESSOR BALDUS. DO YOU KNOW WHAT THAT LOOKS LIKE?
A. WE ACCEPT THE EMENDATION. IT SHOULD BE 18. TYPO.
THE COURT: WHERE?
THE WITNESS: ROW 4, YOUR HONOR, NINE OVER 18. IT’S
OUR MISTAKE. THANK YOU.
MS. WESTMORELAND: YOUR HONOR, ——
MR. BOGER: 1 WAS GOING TD SAY AS AMENDED WE MOVE IT
INTO EVIDENCE.
MS. WESTMORELAND: OTHER THAN THE FACT THAT IT APPEARS
DUE TO THE NUMBERS INVOLVED, I WOULD QUESTION THE SIGNIFICANCE
(OF THE DOCUMENT.
I HAVE NO OBJECTION OTHER THAN THE ONES PREVIOUSLY
NOTED THROUGHOUT.
THE COURT: ILL ADMIT IT.
MR. BOGER: THANK YOU, YOUR HONOR.
rr | sr—_——— ————— m— it
1006
BALDUS - DIRECT
or THE COURT: IM NOT SURE I UNDERSTAND IT.
2 BY MR. BOGER:
3 8: PROFESSOR BALDUS. JUST ONE OR TWO QUESTIONS FOR
4 CLARIFICATION.
WHAT IS THE OVERALL POINT OF THIS TABLE? WHAT DOES (8
THIS TABLE TRY TO SHOW?
"la. THE POINT OF THE TABLE WAS TO TRY AND ESTIMATE THE SOURCES
6
7
8 |OF THE OVERALL RACE OF VICTIM EFFECT THAT APPEAR WHEN WE COMPARE
9 |THE DEATH SENTENCING RATES FOR BLACK AND WHITE VICTIM CASES AMONG
0 |THOSE THAT WERE INDICTED FOR MURDER. THAT PRESENTS THE MEASURE
11 OF THE STATISTICAL EFFECT PRODUCED BY THE ENTIRE SERIES OF
12 DECISIONS IN THE PROCESS.
13 |Q. WHERE DOES IT REFLECT THOSE EFFECTS ARE MOST PROMINENT IN
14 |THE COURSE OF THE CHARGING AND SENTENCING SYSTEM IN FULTON
15 COUNTY?
16 A. THEY APPEAR STRONGEST AT THE. THE PENALTY TRIAL STAGE HERE.
17 THE COURT: IF I SUGGESTED TO YOU THAT MORE THAN HALF
18 |OF THE JURORS IN FULTON COUNTY ARE BLACK, WHAT CONCLUSION WOULD
w 19 |YOU DRAW?
20 |BY MR. BOGER:
21 @. PROFESSOR BALDUS?
22 (A. OH. I THINK THAT’S A FACT, FROM THE BASIS OF WHAT I KNOW,
23 |AND I WOULD CONCLUDE THAT, THAT, WELL IM NOT SURE WHAT I --
24 [CONCLUDE ABOUT WHAT, YOUR HONOR? IT DOESN‘T CHANGE THE NUMBERS,
29 IN TERMS OF THE RATES AT WHICH THE SENTENCE IS IMPOSED. I THINK
1007
BALIUS ~ DIRECT
i
| THAT'S PLAIN.
| BUT I“M NOT QUITE SURE, YOUR HONOR, WHAT I WOULD
|
CONCLUDE WITH RESPECT TO WHAT QUESTION.
| THE COURT: WOULD YOU CONCLUDE THAT THOSE RATES WERE
BEING MADE ON THE BASIS OF DECISIONS PRIMARILY MOTIVATED BY
RACIAL CONSIDERATIONS?
THE WITNESS: 1 WOULD SAY, YOUR HONOR. THAT THE DATA
THAT 1 HAVE HERE THAT I PLACED THE GREATEST RELIANCE ON IS THE.
THE NEAR NEIGHBOR ANALYSIS THAT 1 SHOWED TO YOU EARLIER, AS WELL
AS THE INFORMATION THAT WERE ABOUT TO CONSIDER WITH RESPECT TO
THE POLICE QFFICER CASES.
AND IM NOT SUGGESTING, MY TESTIMONY IS NOT THAT ANYONE
WAS MOTIVATED BY RACIAL CONSIDERATIONS. WHAT 1M SUGGESTING IS
THAT THE CASES ARE PROCESSED DIFFERENTLY. THAT DIFFERENT WEIGHTS
ARE ASSIGNED IN TERMS, IN DETERMINING THE SERIOUSNESS OF THE
CASES IN THE WHITE AND BLACK VICTIM CATEGORIES. THAT'S WHAT THE
EVIDENCE SHOWS, I THINK.
WHY IT SHOWS THAT. THESE DATA DO NOT ADDRESS. YOUR
HONOR. BY MR. BOGER: |
@. PROFESSOR BALDUS. DID YOU OBTAIN ANY INFORMATION ON THE
COMPOSITION OF MR. MCCLESKEY/S OWN JURY, RACIALLY?
A. NO.
THE COURT: DO WE KNOW?
MR. BOCER: I BELIEVE WITH WE‘RE GOING TD HAVE SOME
1008
BALDUS - DIRECT
1 TESTIMONY ON THAT. YOUR HONCR.
2 THE COURT: ALL RIGHT, LET'S TAKE A TEN-MINUTE RECESS.
i hz
4 {RECESS TAKEN.) |
3 THE COURT: GO AHEAD. : ;
i & MR. BOGER: YOUR HONOR, AT THIS TIME, WE INTEND TO
7 OFFER A FINAL REPORT, WHICH WE‘VE ALLUDED TO SEVERAL TIMES
8 DURING THE TESTIMONY OF PROFESSOR BALDUS, WHICH INCLUDES AS A
? PART OF IT A REPORT ON THE FULTON COUNTY CASES.
10 WHAT I WANTED TO INFORM THE COURT OF IS THAT INTENTION
11 TO STATE THAT I WOULD LIKE TO ASK A FEW FOUNDATION QUESTIONS OF
12 PROFESSOR BALDUS, AND I“M AWARE OF YOUR HONOR“S DISINCLINATION
13 NORMALLY TO HAVE TWO PEOPLE QUESTION A WITNESS. I ASK YOUR
14 INDULGENCE ON ANY ARGUMENT OF LAW THAT COMES UP, MR. FORD, MY
15 CO-COUNSEL. HAS DONE A FAIR BIT OF RESEARCH ON WHETHER SUCH A
14 REPORT IS ADMISSIBLE IN EVIDENCE, AND HAS NOT BEEN ABLE TO POUR
17 IT IN MY EARS IN THE TEN MINUTES OR $0 WE HAD ON THE BREAK. 30
13 SINCE ITS NOT A QUESTION OF TWO PEOPLE ARGUING, BUT --
w 1? THE COURT: THATS FINE.
20 MR. BOGER: FINE.
21 tr
22 DAVID C. BALIDUS,
23 BEING PREVIOUSLY DULY SWORN, RESUMED THE WITNESS STAND AND
<4 TESTIFIED FURTHER AS FOLLOWS:
235 DIRECT EXAMINATION (CONT'D)
ho
e
-—
pat
16
1009
BALIUS - DIRECT
BY MR. BUGER?
@. PROFESSOR BALDUS, HAVE YOU COMPLETED ANY DOCUMENT TO REFLECT
THE RESULTS OF YOUR STUDIES AND ANALYSIS, BOTH IN THE CHARGING
AND SENTENCING STUDY, AND THE PROCEDURAL REFORM STUDY?
A. YES.
@. DOES IT CONTAIN AS WELL A REPORT ON THE FULTON COUNTY DATA
THAT 1 ASKED YOU TO EXAMINE?
A. YES, IT IT DOES.
. AND WHO WROTE THAT REPORT?
A. I WROTE THAT REPORT, IN COLLABORATION WITH GEORGE WOODWORTH
AND CHARLES PULASKI.
@. AND HAVE YOU REVIEWED YOURSELF AND VERIFIED ALL THE WRITTEN
AND NUMERICAL CONTENTS OF THAT REPORT?
A. YES.
@. INCLUDING ITS APPENDICES?
lA. YES.
a. AND ARE THE CONTENTS ALL TRUE AND CORRECT TO THE BEST OF
YOUR KNOWLEDGE?
A. YES.
Q. COULD YOU RECALL OR WOULD YOU BE ABLE TO TESTIFY TO THE
| :
(CONTENTS OF ALL THE FACTS AND NUMBERS IN THAT REPORT YOURSELF?
|
|
A. ON THE BASIS OF MY RECOLLECTION. NO.
'@. ALL RIGHT. TO THE BEST OF YOUR KNOWLEDGE, DO THE DATA IN
|THE REPORT ACCURATELY REFLECT THE CONTENTS OF THE RECORDS AND
FILES OF THE GEORGIA DEPARTMENT OF PARDONS AND PAROLES. THE
—— p———_ — ——— —— —— —" —— — —— —————— . rp— yp ———
| 1010
| BALDUS ~- DIRECT
1 GEORGIA DEPARTMENT OF OFFENDER REHABILITATION, AND THE GEORGIA
2 SUPREME COURT, AND THE OTHER DATA SOURCES YOU/VE COLLECTED,
3 |SUPPLEMENTED BY THAT INFORMATION ABOUT WHICH YOUVE TESTIFIED?
3 A. THE UNDERLYING DATA FILE THAT WE USED IN THE ANALYSIS
5 |CORRECTLY REFLECTS THOSE FACTS TO THE BEST OF MY KNOWLEDGE.
& AND THE REPORT THAT WE PREPARED CORRECTLY REFLECTS THE
7 RESULTS OF OUR ANALYSIS OF THOSE DATA.
8 i@. DO YOU KNOW IF COUNSEL FOR THE STATE, THE ATTORNEY GENERAL,
NF] HAS BEEN PUT ON NOTICE THAT THE CONTENTS OF THOSE FILES IN YOUR
10 ANALYSES WOULD BE OFFERED OR BE THE SUBJECT OF PROOF IN THIS
11 CASE?
12 A. ON THE BASIS -- YES.
12 f. AND HOW DO YOU KNOW THAT?
‘A. I°VE BEEN ADVISED BY COUNSEL TO THAT EFFECT. AND ALSO THAT
15 en A MATTER THAT WAS IMPLICIT IN THE DEPOSITION THAT WAS
16 CONDUCTED IN SYRACUSE.
17 #2. THAT WAS THE DEPOSITION OF YOU?
18 A. YES.
Ww 19 QR. BY THE ATTORNEY GENERAL?
20 A. YES,
23 QR. AND HAVE YOU TURNED OVER TO THE STATE AT ANY POINT THROUGH
22 COUNSEL COPIES OF YOUR FINAL REPORT WITH AFPENDICES?
=4 @. LET ME ASK YOU FURTHER. TO YOUR KNOWLEDGE IS IT POSSIBLE FOR
23 A CITIZEN OR ANYONE TO OBTAIN COPIES OF THE OFFICIAL REPORTS OF
po
y
MN
Ww
{
{
|
1011
BALDUS - DIRECT
THE GEORGIA DEPARTMENT OF PARDONS AND PAROLES. TO WALK OUT OF
THE OFFICE WITH THEM AND USE THEM AS ONE CHOOSES?
A. NO, IT’S NOT POSSIBLE.
@. AND HOW DO YOU HAVE THAT KNOWLEDGE?
A. BECAUSE I WENT THROUGH THE PROCESS OF TRYING TO GAIN ACCESS
TO THOSE FILES. WHICH I DID, AND IT REQUIRED THE APPROVAL OF THE
BOARD OF PARDONS AND PAROLES.
GB. AND WHEN YOU OBTAINED ACCESS TO THEM. WERE YOU GIVEN LEAVE
FREELY TO TAKE THOSE DQCUMENTS THEMSELVES. AND BRING THEM
SOMEWHERE OR TAKE THEM ANYWHERE YOU CHOSE OR REPRODUCE THEM?
A. NO. THE CIRCUMSTANCES OF OUR ACCESS TO THOSE DATA WAS
STRICTLY LIMITED TO THE COLLECTION OF DATA ON THE PREMISES, THE
ENCODING OF THE INSTRUMENTS, AND THE USE OF THEM STRICTLY IN
ANALYSIS RELATED TO OUR RESEARCH.
@. COULD YOU DESCRIBE TO THE COURT BEYOND THE ANALYSES WHICH
YOU HAVE TESTIFIED TO TODAY. THE VOLUME OF ANALYSES THAT ARE
REFLECTED IN THAT FINAL REPORT?
A. THE FINAL REPORT REFLECTS A SUBSTANTIAL AMOUNT OF ANALYSIS
WHICH HAS BEEN ONGDING FOR APPROXIMATELY ONE YEAR. AND ACTUALLY MORE THAM A YEAR. AND I HAVE CONDUCTED AN EXTENSIVE AMOUNT OF
ANALYSIS, WHICH IS DISTILLED INTO THIS REPORT.
@. WE’VE PUT IN SOME, I THINK, FORTY-FIVE OR SQ GRAPHS OR
FIGURES AND TABLES. ARE THERE ADDITIONAL GRAPHS AND FIGURES AND
TABLES IN YOUR FINAL REPORT BEYOND THAT?
A. YES, THERE ARE A SUBSTANTIAL NUMBER PLUS A SUBSTANTIAL BODY
1012
BALDUS ~ DIRECT
1 OF TEXT WHICH DESCRIBES THE METHODOLOGY AND THE CONCLUSIONS IN
2 FULLER DETAIL THAN I HAVE PRESENTED HERE.
3 @. AND DO YOU RELY ON THOSE ADDITIONAL TABLES AND FIGURES AND
4 GRAPHS WITH THE METHODOLOGY AS SET FORTH IN PART AS A BASIS FOR
3 THE EXPERT OPINIONS THAT YOU’VE DELIVERED?
® Fr ow
7 R. IN THIS COURT?
8 A. YES.
RJ
@. HAVE ALL YOUR UNDERLYING RECORDS FROM WHICH THIS REPORT WAS
10 MADE, YOUR DATA PRINTOUTS, AND COMPUTER ITEMS AND SO FORTH. BEEN
11 MADE AVAILABLE FOR INSPECTION AND COPYING BY THE STATE?
12 A. YES.
13 2. WHEN?
i4 A. IN». IN LATE JUNE OR EARLY JULY OF THIS YEAR. I DON’T RECALL
15 THE EXACT DATE.
16 @. AND ARE YOU WILLING TO, NOW TO PRODUCE YOUR RECORDS IF
17 REQUESTED BY THE STATE OR BY THE COURT?
13 A. YES.
w 1% MR. BOGER: EXCUSE ME, ONE MINUTE, YOUR HONOR.
20 BY MR. BOCER:
21 R. 1S THERE ANY PRACTICAL WAY BY WHICH YOU COULD BRING ALL
22 THOSE INTO THIS COURTROOM AND TESTIFY AS TO THEM WITHOUT THE USE
23 OF A REPORT IN ANY REASONABLE PERIOD OF TIME, AND I AM TALKING
24 NOW ABOUT A PERIOD LIKE A WEEK OR TWO?
23 A. NO, NOT THAT I KNOW OF,
1013
BALDUS -~ DIRECT
1 MR. BOBER: YOUR HONOR, AT THIS POINT, I BELIEVE WE
|
<2 WOULD MOVE THE ADMISSION INTO EVIDENCE OF PROFESSOR BALDUSC
3 FINAL REPORT. INCLUDING A FULTON COUNTY REPORT AND THE
4 APPENDICES, AND AS I HAVE INDICATED, MR. FORD WOULD ADDRESS ANY
3 LEGAL QUESTIONS THAT MIGHT ARISE ON THIS MATTER.
Ri & THE COURT: I WOULD LIKE TO SEE ONE BEFORE THE ARGUMENT
7 BEGINS.
8 MR. BOGER: I HAVE ONE AVAILABLE FOR THE COURT.
? THIS IS THE REPORT ITSELF.
10 THESE ARE THE APPENDICES.
11 COULD WE MARK THESE FOR IDENTIFICATION AS DB-113, YOUR
12 HONOR?
13 THE COURT: ALL RIGHT.
14 MR. BOGER: PERHAPS DB-113A AND -B, SINCE THERE ARE TWO
15 VOLUMES.
14 THE COURT: ALL RIGHT, MR. FORD.
17 MR. FORD: THANK YOU, YOUR HONOR.
13 AT THE OUTSET I WOULD LIKE TO MAKE IT CLEAR WHAT OUR
w 17 PROFFER 1S HERE.
20 THE REPORT IS AN EXTENSIVE DISCUSSION OF THE STUDY AND
21 ANALYSIS AND SUMMARY OF PROFESSOR BALDUS” FINDINGS, CONCLUSIONS.
22 METHODS, AND THEIR BASIS.
23 | WE OFFER IT AS SUCH. AND AS SUCH, I THINK THE COURT
24 HAS ADMITTED MANY DOCUMENTS THAT ARE ULTIMATELY CONTAINED IN IT
23 ALONG THE WAY THROUGH HIS TESTIMONY.
1014
BALDUS - DIRECT
1 WE ALSO OFFER IT AS SUBSTANTIVE EVIDENCE OF THE FACTS
2 ASSERTED THEREIN. THE FACTS OF THE CASES, THE ACTUAL FACTS OF
3 THE CASES CONTAINED IN THE STUDY: THE ACTUAL FACTS KNOWN OR
4 BELIEVED BY THE DECISION MAKERS IN THOSE CASES: THE FACTS OR THE
3 CONTENTS OF THE OFFICIAL RECORDS FROM WHICH THE STUDY DATA WERE
& DERIVEDS AND THE FACTS, AS WELL AS THE FACTS CONSIDERED AND
7 ANALYZED BY PROFESSOR BALDUS IN REACHING HIS FINDINGS AND HIS
8 CONCLUSIONS.
2 NOW, OF COURSE, ON NONE OF THESE SCORES IS THIS
10 EVIDENCE, LIKE ANY OTHER EVIDENCE, CONCLUSIVE. AND LIKE ANY
31 OTHER EVIDENCE, ITS WEIGHT WITH REGARD TO ANY OF THESE QUESTIONS
12 IS SUBJECT TO THE COURT’S ULTIMATE CONSIDERATION OF ITS
13 CREDIBILITY, THE METHODS BY WHICH IT WAS OBTAINED AND ALL THE
14 MATTERS THE COURT HAS HEARD THAT RELATE TO THOSE IN EACH AREA.
15 ALSO, I THINK THE LEGAL SIGNIFICANCE OF EACH LEVEL OF
16 THOSE FACTS. AND THE FACTS RELATED IN THE REPORTS, IS A MATTER
$7 OF COURSE FOR THE COURTS ULTIMATE DETERMINATION WHEN IT REACHES
13 ITS CONCLUSIONS AS TO THE LAW GOVERNING THIS CASE.
A 12 BUT WE BELIEVE THAT ITS IMPORTANT THAT THE RECORD
20 REFLECT THAT THESE DATA ARE PROPERLY BEFORE THE COURT FOR ITS
21 CONSIDERATION IN ALL THOSE REGARDS. INSOFAR AS IT FINDS THEM
22 CREDITABLE AND SUBSTANTIVE AND PERSUASIVE.
23 NOW BEFORE I TURN TO THE RULES OF EVIDENCE WHICH
24 OBVIOUSLY ARE GOING TO GOVERN THIS QUESTION AS TO WHETHER THESE
29 ARE ADMISSIBLE, THERE ARE A COUPLE OF PRACTICAL CONCERNS.
1015
BALDUS - DIRECT
1 CERTAINLY INSOFAR AS IT REFLECTS PROFESSOR BALDUS”
<2 ANALYSIS, THERE IS A MATTER OF CONVENIENCE. THIS REPORT DOES
3 SUMMARIZE AND MAY BE MUCH MORE AVAILABLE TO THE COURT AS IT
CONSIDERS THIS CASE THAN WOULD BE THE TRANSCRIBED RECORD OR
PERHAPS UNTRANSCRIBED RECORD.
THERE ALSO, FRANKLY, IS A CONSIDERATION OF
4
bo
5
J COMPLETENESS. THE COURTS INDICATED THAT IT HAS A CONCERN THAT
8 THIS RECORD BE DEFINITIVE OF AT LEAST WHAT IS KNOWN AT THIS
POINT, WHICH WE BELIEVE IS A STATE OF THE ART DETERMINATION OF
0 THESE QUESTIONS.
1! CERTAINLY THAT IS OUR CONCERN, THAT THROUGH THE
12 COMPLEXITY OF THIS, THROUGH THE LIMITATIONS OF TIME TO COVER
13 SOMETHING THIS MASSIVE, WE HAVEN’T LOST SOMETHING.
14 THERE ALSO IS SIMPLY THE PRACTICAL FACT IF WE WERE TO
13 GO BACK AND SIMPLY ASK PROFESSOR BALDUS TO RE-REVIEW EVERY
16 SINGLE FACT THAT CONTRIBUTES AND GET IT SUBSTANTIVELY IN, IT
17 WOULD SIMPLY BE IMPOSSIBLE.
18 NOW THERE ARE RULES OF EVIDENCE IN THE FEDERAL RULES,
. | 1? AND WE“VE DONE SOME STUDYING SINCE THE COURT POINTED QUT THAT
| 20 THERE IS A DIFFERENCE BETWEEN A MATTER BEING CONSIDERED BY AN
21 EXPERT AND A MATTER BEING IN EVIDENCE, THERE ARE SPECIFIC RULES
22 OF EVIDENCE WHICH I BELIEVE ADDRESS THE LEGAL QUESTIONS THAT
23 CONTROL ON EACH ONE OF THESE SCORES. AND I THINK THE RECORD IS
24 ABSOLUTELY CLEAR THAT SUBJECT TO THE DETERMINATIONS OF WEIGHT.
23 SUBJECT TD FINAL CONSIDERATION OF RELEVANCE. FOR THE. OF EACH
1016
BALDUS - DIRECT
1 AND EVERY FACT ANWYAY, THESE ARE ADMISSIBLE EVIDENCE.
Ps NOW TO START, 1 BELIEVE ITYS IMPORTANT, BECAUSE AT THE
3 FIRST LEVEL WERE TALKING ABOUT THE DIFFERENCE BETWEEN THE
4 ACTUAL FACTS, THE FACTS KNOWN BY THE DECISION MAKERS, AND THE
or’ FACTS REFLECTED IN THE OFFICIAL RECORDS, I THINK IT’S IMPORTANT
8 & TO BEGIN WITH RULE OF EVIDENCE 803(8).
7 AND I PROFFER TO THE COURT AT THIS TIME, AND ASK THE
COURT TO TAKE JUDICIAL NOTICE OF GEORGIA STATUTE SECTION 77-012
G
d
0
REGARDING THE OBTAINING OF INFORMATION BY THE GEORGIA DEPARTMENT
10 OF PARDONS AND PAROLES RELATIVE TO THESE CASES. I THINK IT’S
11 HIGHLY SIGNIFICANT ON THIS QUESTION. COULD I --—
12 THE COURT: WHAT DOES IT SAY?
13 MR. FORD: IT SAYS, YOUR HONOR. THAT IT IS THE DUTY OF
14 |THE ROARD TO OBTAIN AND PLACE IN ITS PERMANENT RECORDS
135 INFORMATION AS COMPLETE AS MAY BE PRACTICABLE ON EVERY PERSON
14 WHO MAY BECOME SUBJECT TO ANY RELIEF WHICH MAY BE WITHIN THE
17 POWER OF THE BOARD TO GRANT. |
12 THE COURT: LETS CUT THROUGH ALL THIS MR. FORD.
W 19 THIS IS AN OUT-OF-COURT STATEMENT OFFERED TO FROVE THE
20 TRUTH OF THE MATTERS ASSERTED THEREIN.
<1 "EVEN IF I GIVE YOU RELEVANCY AS TO THE ENTIRETY OF IT,
22 WHICH FOR THE PURPOSES OF THESE ANALYSES. 1 WILL DO, YOU HAVE
23 NOTED A LOT OF THE REASONS, POLICY REASONS THAT WOULD MILITATE
24 TOWARD MY LETTING IT IN, AND I“M SURE THE STATE“S GOT SOME OF
23 ITS OWN, AND SO DO I HAVE SOME POLICY REASONS FOR GOING THE
1017
BALDUS - DIRECT
1 OTHER WAY.
BUT LET ME SEE IF YOU HAVE THOUGHT OF HOW THIS REALLY §
3 BECOMES ADMISSIBLE, ASSUMING ITS RELEVANCY.
4 MR. FORD: IF I CAN JUMP AHEAD, THE COURT S ASKED ME TO
JUMP AHEAD FROM THE OFFICIAL RECORDS QUESTION. 4]
WITH REGARD TO THE SUMMARY ITSELF, I THINK THE
& .
PRINCIPLE, THERE“S 3 BASIC RULES THAT I THINK APPLY HERE --
WELL, THERE‘S TWO THAT APPLY TO THE SUMMARY.
g
o
.
0
N
D
ONE IS RULE OF EVIDENCE 1004, THE OTHER IS RULE OF
10 EVIDENCE 10046. I THINK THEY BOTH ADDRESS EXACTLY THIS QUESTION,
11 AND =~
12 THE COURT: SUMMARIES?
13 MR. FORD: ONE IS ON ADMISSIBILITY OF QTHER EVIDENCE OF
14 THE CONTENTS OF A WRITING. THAT IS IN SPECIFICALLY RULE 1004(3)
1% WHICH DEALS WITH EVIDENCE OF THE CONTENTS OF WRITINGS THAT ARE
16 IN THE POSSESSION OF THE PARTY AGAINST WHOM IT IS OFFERED. WHEN
17 THE PARTY HAS BEEN PUT ON NOTICE THAT THE CONTENTS CF THOSE
18 RECORDS WOULD BE THE SUBJECT OF PROOF AT A HEARING, AND THE
K's 1? OTHER PARTY DOES NOT PRODUCE THE ORIGINALS AT THE HEARING.
20 | THE SECOND IS RULE 1006, WHICH DEALS WITH THE CONTENTS
21 OF VOLUMINOUS WRITINGS AND RECORDINGS WHICH CANNCOT BE
22 CONVENIENTLY EXAMINED IN COURT. AND WHICH MAY THEREFORE BE
23 PRESENTED, I‘M NOT READING EXACTLY, I“M PARAPHRASING. IN THE
24 FORM OF A CHART SUMMARY OR CALCULATION, WHERE THE ORIGINALS ARE
235 MADE AVAILABLE AND THE COURT MAY ORDER THEM PRODUCED IN COURT IF
1013
BALDUS - DIRECT
Fe
y IT WISHES.
rn
I BELIEVE THAT THOSE ARE THE CONTROLLING LAW. THERE IS
|
'A FIFTH CIRCUIT CASE ~-—
THE COURT: LET ME TELL YOU WHERE YOU’VE CONTROLLED
YOURS TO. 1 MIGHT VERY WELL SAY, OKAY, THAT’S FINE, COUNSEL.
$ STRIP THIS DOCUMENT OF EVERY PARTICLE OF ANALYSIS AND THEN I
WILL CONSIDER IT UNDER THOSE TWO.
lod DOESN’T GET THE ACCOMPANYING, NEITHER ONE OF THOSE GET
|THE ACCOMPANYING ANALYSIS INTO EVIDENCE.
N
y
«w
w
Oo
1
&
W
L
t
MR. FORD: IT SEEMS TO ME. YOUR HONOR, THAT THE
11 ANALYSIS IS AGAIN. WELL, ANALYSIS, I THINK WOULD CONCLUDE ALL
i2 THE RUNS THAT PROFESSOR BALDUS HAS MADE AND ALL THE -——
13 THE COURT: ASSUMING FOR THE SAKE OF ARGUMENT THAT ANY
14 NUMBER IN THAT BOOK IS ADMISSIBLE UNDER SOME THEORY OF EITHER
13 1004 OR 1004. THAT DOESN‘T GET THE EXTENSIVE, THAT'S WHY I
ié6 WANTED TO LOOK AT A COPY OF IT, THE EXTENSIVE TESTIMONIAL
17 ANALYSIS INTO EVIDENCE,
13 MR. FORD: THERE ARE, 1 GUESS MY QUESTION THERE WOULD
w 1? BE WHETHER THERE’S ANY OBJECTION ON THAT GROUND, BECAUSE IT |
20 SEEMS TO ME THAT THAT KIND OF PURPOSE HAS BEEN, FOR THAT KIND OF
<1 PURPOSE, MUCH OF THIS HAS BEEN ADMITTED FOR PURPOSE OF
22 REFLECTING THOUGHT PROCESSES AND ANALYSIS, IT HAS BEEN ADMITTED.
rac AND EVEN MUCH OF IT WITHOUT OBJECTION, FOR THAT PURPOSE.
24 THE COURT: TALKING ABOUT NUMBERS.
25 MR. FORD: THE NUMBERS SEEM TO ME TO BE THE HARD
1012
BALDUS - DIRECT
1 QUESTION YOUR HONOR, THE QUESTION THE COURT'S NOT YET REACHED.
2 THE COURT: NO, WHAT 1 HAVE BEEN LETTING IN IS NUMBERS
3 AS THEY REFLECT PROCESS. AND I HAVE LET SOME OF IT IN.
4 BUT LET ME TELL YOU FRANKLY WHAT MY DUAL CONCERNS ARE.
= AND THEY ARE THE OTHER SIDE HAS TWO ADVANTAGES THAT YOU HAVE
b POINTED QUT.
7 - AFTER I HAVE HEARD WHAT I HAVE HEARD THIS AFTERNOON, I
3 AM AT A POINT WHERE I REALIZE I HAD STARTED ON A LABOR OF LOVE
2 FRANKLY FOR YOU IN THIS CASE, AND I REALIZE THAT MY LABOR HAS
0 BEEN MORE LOVING THAN 1 THOUGHT.
11 WE MAY BE A LONG WAY AWAY FROM YOUR DEMONSTRATING UNDER
12 THE TESTS IN THIS CIRCUIT ANY STATISTICAL EVIDENCE THAT MIGHT
13 GET YOU A GRANT OF WRIT OF HABEAS CORPUS FOR MCCLESKEY., BECAUSE
14 A LOT OF GOOD RESEARCH HAS GONE ON IN THIS AREA, AND DOCTOR
13 BALDUS IS OBVIOUSLY A WELL RUALIFIED PERSON, I HAVE TRIED TO LET
14 YOU MAKE A FULL RECORD FOR EVERY PURPOSE. NOT JUST FOR THE
17 PURPOSES OF THIS CIRCUIT. BUT FOR EVERY PURPOSE.
18 THAT RECORD OUGHT TO BE A LAWYER‘S RECORD, NOT AN
Ww 1? ACADEMICIAN’S RECORD. THE PERSON THAT REVIEWS THIS RECORD OUGHT
20 TO HAVE TO REVIEW THAT TRANSCRIPT OF WHAT THIS MAN SAID UNDER
21 OATH IN OPEN COURT, SUBJECT TO CROSS-EXAMINATION.
22 | NOW, THERE ARE A BUNCH OF REASONS FOR THAT. THE UNDER
23 0aTH, I THINK, IS OBVIOUS. THE WAY HE TESTIFIED IS WHAT HE IS
24 THINKING WHERE HE IS NOW. IT IS HIS THOUGHT.
25 IT IS THE STRONG NOTICE TO COUNSEL OF WHAT THE ISSUE IN
| 1020
BALIWIS - DIRECT
THE CASE IS.
WHAT YOURE ASKING ME TO DO IS TO LET THIS IN AS A
SAFETY NET FOR THINGS YOU-VE MISSED.
THE OTHER SIDE OF THE COIN IS THAT IF WE DON‘T LITIGATE
THAT ISSUE AS TO ESSENTIALLY ORAL TESTIMONY IN THIS COURT.
COUNSEL HASNT HAD NOTICE OF WHAT ISSUE MIGHT BE LURKING IN
THERE THAT BECAUSE IT DIDN'T COME OUT IN OFEN COURT, SHE DIDNT
HEAR THE WITNESS TESTIFY TO IT. AND SHE DOESN‘T KNOW WHETHER
IT’S THERE OR NOT.
HAVING GLANCED OVER THIS, I THINK I WILL PUT THIS
ARGUMENT TO A QUICK END BY TELLING YOU THAT I AM NOT GOING TO
ADMIT IT. I WILL ALLOW YOU TO PROFFER IT, SUBJECT TO THE
CONDITIONS THAT YOU PREVIOUSLY PROFFERED IT. 1 AM GOING TO SEAL
IT. AND I PRAY TO GOD THAT NO LAWYER THAT HAS TO LOOK AT THIS
RECORD OR EVER OPENS IT, BECAUSE I THINK -— I UNDERSTAND WHY
YOURE DOING IT, AND IF I WERE IN YOUR SHOES I WOULD GO DO THE
SAME THING. SO I“M NOT BEING CRITICAL OF YOU AS A LAWYER. BUT
1 CANT IMAGINE A MORE SERIOUS INTERVENTION TO THE WEIGHING AND
EVALUATING OF EVIDENCE THAN TO BASICALLY SUBSTITUTE A HEARSAY
STATEMENT FOR TESTIMONY UNDER OATH SUBJECT TO CRNSS-EXAMINAT ION
THAN TO ALLOW A BOOK LENGTH SUMMARY OF THE WITNESS” TESTIMONY
IN.
AND IT IS FOR THOSE ESSENTIAL POLICY REASONS I WILL NOT
LET IT IN BEYOND THE FACT THAT IT IS NOT ADMISSIBLE AS A HEARSAY
STATEMENT, AND THATS THE PRINCIPAL PURPOSE DF 17.
20
1021
BALDUS ~ DIRECT
IF THERE ARE SOME SUMMARIES OF RECORDS AND DOCUMENTS
THAT YOU WANT IN, I THINK I HAVE BEEN LETTING A LOT OF THOSE IN.
TO THE EXTENT THAT THERE®S AN ANALYSIS IN THERE THAT
DEMONSTRATES ANY FACT OF CONSEQUENCE IN THE CASE, 1 HAVE BEEN
LETTING THOSE IN. SO I DON’T THINK YOU ARE AT ALL DISADVANTAGED
BY ME NOT LETTING YOUR RAW DATA OR YOUR SUMMARIES OF YOUR RAW
DATA OR IN SOME CASES. WHERE SOME SOPHISTICATED ANALYSIS,
SOMETHING BEYOND JUST PLAIN OLD HIGH SCHOOL. ALGEBRA. OR, I DON'T
EVEN KNOW WHETHER IT WOULD BE HIGH SCHOOL ALGEBRA. KIND OF
BEYOND THE BOWERS-PIERCE LEVEL, WHERE SOMETHING HAS BEEN DONE TO
SUGGEST THAT THIS IS MORE THAN JUST A PURE NUMBERS GAME. I HAVE
LET THOSE IN TO PROVE THE PROOF OF THE MATTERS ASSERTED THEREIN.
SO TO THE EXTENT THAT YOU HAVE BURDENS WITH DATA.
EITHER IN TERMS OF GETTING IN THE UNDERLYING THING, I“LL BE GLAD
TO HEAR FROM YOU ABOUT YOUR CONCERNS. TO THE EXTENT YOURE
HAVING TROUBLE GETTING IN ANALYSIS. AS IT BEARS ON THE ISSUE,
ILL BE GLAD TG DISCUSS THOSE THINGS WITH YOU. IN OTHER WORDS,
AS TO NUMBERS. AND NUMBERS THAT RESULT FROM ANALYSIS, THATS ONE
THING. BUT THAT TESTIMONY OR EVIDENCE IS IN NO FASHION GOING
INTO THIS RECORD.
MR. FORD: MAY I THEN ADDRESS A SOMEWHAT DIFFERENT
SUBJECT, THIS MAY BE KIND OF THE FLIP SIDE OF THIS.
AND IN MY, WHAT, THE LAST FEW DAYS, A NUMBER OF ITEMS
HAVE COME INTO EVIDENCE FOR LIMITED PURPOSES. THERES BEEN A
NUMBER OF OBJECTIONS MADE WHICH I“VE NOT CLEARLY UNDER3TOOD. IM
rr ———————— —————
1022
BALDUS ~ DIRECT
1 NOT SURE THE COURTS CLEARLY UNDERSTOOD. THEY “VE NOT BEEN
2 hl sh ON ANY RULE OF EVIDENCE, AND THE COURT HAS AT TIMES
LIMITED CERTAIN THINGS TO AN EVIDENTIARY, DIFFERENT EVIDENTIARY
RULES. AND DOCTOR BALDUS HAS ALSO SPOKEN ORALLY OF CERTAIN
NUMBERS BUT NOT OTHERS.
3
4
5
ks 6 WOULD THE COURT, 1 UNDERSTAND THE COURT’S RULING WITH
7 REGARD TO THE REPORT ITS, S0 THAT WE COULD MAKE A DECISION AT
3 THIS POINT: WITH REGARD TO ANY HEARSAY-TYPE OBJECTIONS,
9 PROCEDURAL~-TYFE OBJECTIONS, AND OF COURSE RESERVING THE ULTIMATE
10 QUESTION OF RELEVANCY AND WEIGHT. WOULD THE COURT ADMIT
11 SUBSTANTIVELY THE NUMBERS, THE DATA, THE FACTS THAT PROFESSOR
12 BALDUS HAS SPOKEN OF, AND THAT ARE CONTAINED IN THE EXHIBITS
13 ‘WHICH HAVE BEEN ADMITTED, FOR SOME PURPOSE?
14 | THE COURT: IF YOURE TALKING ABOUT RAW DATA: THAT MAY
15 BE ONE THING. IF YOUYRE TALKING ABOUT ANALYSIS, THAT MAY BE
16 ANOTHER THING. AND I WILL HAVE TO SIT DOWN AND SEE WHAT YOUR
17 SPECIFIC CONCERNS ARE.
12 NOW, LET ME CLEAN UP THE RECORD.
w 12 AN AWFUL LOT OF WHAT WE-“VE BEEN DOING HERE I5 TO
20 DEMONSTRATE THAT THIS IS AN INTELLECTUALLY HONEST UNDERTAKING,
21 AND LOT OF THE BUILDUP A3 MR. BOGER HAS PRESENTED IT TO ME, HAS
22 TAKEN ME STEP BY STEP. THE ACADEMICIAN, LOOKING AT SOME
<3 FIGURES, SAYING WHAT DOES THIS SUGGEST. AND THEN IF I LOOK AT
24 HIS TESTIMONY, AND HE HAS TESTIFIED. I1°VE NOTICED THIS. BECAUSE
23 I HAD TO JUDGE THE BOWERS-PIERCE TESTIMONY, HE“S TESTIFIED ANY
S
H
O
R
E
S
E
E
RE
a
D
O
10
11
| 1023
BALDUS - DIRECT
NUMBER OF TIMES IN THIS HEARING THAT BASED ON THESE EARLY PIECES
OF ANALYSIS, HE WOULD NOT GIVE AN EXPERT OPINION THAT THIS WAS
HAPPENING.
| WHAT I“M SAYING IS, THAT DOESN‘T PROVE, THAT EARLY
ANALYSIS. THAT RAW FIGURE, GEE WHIZ, THERE'S A DISPARITY SORT OF
STUFF. YOUR OWN WITNESS HAS SAID DOESN‘T REALLY PROVE IT, USING
THE BEST TECHNIQUES OF EITHER STATISTICS OR SOCIAL SCIENCE.
THEREFORE, IT DOESN’T A MAKE A FACT CONSEQUENCE MORE OR
LESS LIKELY IN THIS CASE, AND IT’S IRRELEVANT.
NOW, 1 DON’T KNOW WHAT RAW DATA YOU/RE TRYING TO GET
IN. BUT IF YOU WILL SHOW ME WHAT RAW DATA YOU NEED IN THE
RECORD, ILL BE GLAD TO LET YOU PRESENT IT IN A CONVENIENT
SUMMARY FASHION. BUT I AM NOT GOING TO LET YOU PUT INTO THE
RECORD UNDER THAT GUISE ANALYSIS THAT COULD BE MISINTERPRETED BY
ANYONE IN OR OUT OF THE JUDICIAL SYSTEM.
MR. FORD: YOUR HONOR. PERHAPS IF I COULD JUST SPEAK
AGAIN TO LET YOUR HONOR KNOW THE CONCERNS WE HAVE.
MY DOUBT IS. I THINK THE COURT’S BEEN VERY CLEAR ON THE
RELEVANCE RULINGS AND I‘M NOT MEANING TO ARGUE THOSE AGAIN, I
THINK WE‘VE MADE OUR POSITION CLEAR AND THE COURT/S MADE ITS
POSITION CLEAR ON THAT ISSUE.
BUT WITH REGARD TO WHAT NUMBERS ARE HERE AS REFLECTING
A REAL WORLD, A REAL STATE. AND A REAL DECISION MAKER, WITH
WHATEVER SIGNIFICANCE THE COURT FINDS THEY HAVE, GIVEN THE
MANNER IN WHICH THEY WERE GATHERED, I“M NOT SURE WHAT S IN AND
+
Sn
el
1 1
BE
R
FO
RE
i
10
33
1024
BALDUS - DIRECT
WHATS OUT, OF PROFESSOR BALDUS- TESTIMONY. AND OF THESE
EXHIBITS. AND WHAT IM AFRAID OF, IF THE COURT HAS MADE
RULINGS ON ONE EXHIBIT IN WHICH PROFESSOR BALDUS READ FOUR
NUMBERS WHICH WERE NOT OBJECTED TO. BUT THE COURT THEN SAID THIS
IS IN FOR ANALYSIS ONLY, BUT NOT FOR FACTS, AS I UNDERSTAND THE
LAW, COUNSEL CAN THEN MAKE ANY EVIDENTIARY OBJECTION THEY WANT
ON APPEAL TO SUPPORT YOUR HONORS RULING IF THAT’S EXCLUDED. AND
WERE GOING TO END UP WITH A FACTUAL RECORD THAT LOOKS LIKE |
SWISS CHEESE.
WITH REGARD TO DATA, WHAT I WOULD ASK THE COURT. AT
LEAST SO WE CAN DETERMINE OUR POSITION. I WOULD ASK THE COURT
THEN AT THIS TIME, TO, IS THE COURT SAYING THAT THE DATA THAT'S
DISCUSSED BY PROFESSOR BALDUS AND BY, IN THOSE DOCUMENTS THAT
THE COURT HAS RULED ARE RELEVANT, IS THAT DATA BEFORE THE COURT
‘AT THIS TIME, FOR SUBSTANTIVE PURPOSES, TO PROVE THESE ULTIMATE
FACTS?
THE COURT: I THINK THE ANSWER IZ NO. BUT I HAVENT
THOUGHT THAT THROUGH.
I WOULD THINK, MR. FORD, WITHOUT REFLECTING FURTHER,
THAT TO THE EXTENT THAT I HAVE INTRODUCED THE DOCUMENTS. NOT
INTRODUCED, ADMITTED THE DOCUMENT FOR PROOF OF THE MATTER
ASSERTED THEREIN. THAT THE DATA CONTAINED IN THAT DOCUMENT IS IN
THE RECORD. IT IS IN EVIDENCE AND ONE MAY ARGUE WITH IT AS ONE
SEES FIT.
TO THE EXTENT THAT IT IS A SUMMARY OF A VOLUMINOUS
Ww
0
s
BALDUS - DIRECT
RECORD, WRITING OR RECORDING, THEN EVERY PIECE OF DATA WRITTEN
DOWN ON IT AND THE FAIR INFERENCES FROM IT ARE IN EVIDENCE.
TO THE EXTENT THAT I HAVE EXCLUDED ANY DF THESE THAT
DID NOT CONTAIN SOPHISTICATED ANALYSIS OR THE ANALYSIS IS NOT IN
EVIDENCE. THE DOCUMENT IS NOT IN EVIDENCE. I HAVE ORDERED THOSE
DOCUMENTS WHICH I HAVE EXCLUDED SEALED. THE DOCUMENT IS ONLY IN
TO SHOW PROCESS, NOT TO SHOW FACTS.
IM JUST NOT SURE OF WHAT PIECE OF DATA YOU WANT IN THE
RECORD THAT IS NOT IN THE RECORD.
MR. FORD: YOUR HONOR, ILL HAVE TO BE FRANK. I“VE
NEVER BEEN IN A SITUATION BEFORE IN COURT IN WHICH DOCUMENTS ARE
MARKED ADMITTED, BUT THEY“RE NOT ADMITTED FOR PURPOSES OF FACTS
ASSERTED. EXCEPT BEFORE A JURY, WHERE THE INSTRUCTIONS MAKE
CLEAR WHAT IS IN FOR WHAT PURPOSES. AND THAT'S WHY I THINK
NE-“RE we
THE COURT: HAVING HEARD THE WHOLE THING, MR. FORD.» IF
YOU WISH TO PRESS IT, I WOULD BE GLAD NOW, SINCE I HAVE HEARD
THE TESTIMONY. AND SINCE I NO LONGER NEED THE EXHIBIT, TO TELL
YOU THAT I NO LONGER NEED THOSE THAT I HAVE ADMITTED FOR
PROCESS, AND WILL BE GLAD TO EXCLUDE THEM ALL, AND THEN THERE
WON“T BE ANY CONFUSION. 1 WAS TRYING TO BE GENEROUS WITH THOSE
RULINGS, AND YOURE TRYING TO MOUSETRAP ME WITH THEM.
I DONT NEED THEM. I UNDERSTAND NOW WHAT HE HAS DONE,
AND UNDERSTANDING WHERE WE ARE NOW, I DON’T NEED THEM.
THEY ARE A BURDEN ON THE RECORD AND IF THEY ARE A
1026
BALDUS - DIRECT
1 BURDEN TO YOU I WILL EXCLUDE THEM. 2 MR. FORD: THE ONLY BURDEN IM ASKING RELIEF FROM IS, I
THINK I“M GETTING RELIEF FROM. IS UNCLARITY. IF I UNDERSTAND
CORRECTLY, THEN, PROFESSOR BALDUS” TESTIMONY INSOFAR AS ITS
BEEN ADMITTED WITHOUT OBJECTION WITH REGARD TO DATA IS BEFORE
3
4
5
& & THE COURT SUBSTANTIVEVLY, AND THOSE DOCUMENTS THAT HAVE BEEN
7 ADMITTED SUBSTANTIVELY ARE BEFORE THE COURT SUBSTANTIVELY, AND
3 OTHER THINGS ARE NOT. IS THAT A CORRECT UNDERSTANDING OF THE
E COURTS RULING AT THIS POINT?
10 THE COURT: I THINK IT WOULD FOLLOW, MR. FORD, THAT IF
11 I SAID THAT THE ANALYSIS AND THE DOCUMENT IS NOT RELEVANT, THAT
|
12 TESTIMONY. READING THE DOCUMENT IS LIKEWISE NOT RELEVANT. 13 NOW, WE’RE TRYING THIS CASE NONJURY., AND WHAT I THINK I
14 HAVE DONE IS TO LET YOU PUT IN THE DOCUMENTS THAT DRAW THE
13 (CONCLUSIONS, BASED ON MORE SOPHISTICATED SOCIAL SCIENCE AND
14 STATISTICAL TECHNIQUES. THOSE ARE IN EVIDENCE TO SHOW WHAT THEY
17 SHOW.
i3 1 DON’T UNDERSTAND THE STATE TO HAVE SERIOUSLY OBJECTED
Ww 19 THAT THEY MISREPRESENT THE STATE OF THE DATA ON WHICH THEY ARE
20 COMPILED. IF THEY DQ. THEN YOU‘VE COT A REBUTTAL CASE WHEN YOU
<1 CAN DO SOMETHING ABOUT IT.
22 I DON’T UNDERSTAND THE PROBLEM. AND YOU MAY NOT BE
23 TELLING IT, YOU MAY EITHER BE TRYING TO GET SOMETHING IN THE
24 RECORD FOR ANOTHER PURPOSE THROUGH THIS GUISE OR YOU MAY BE
22% WORRYING ABOUT HAVING SOMETHING IN THE RECORD AND YOURE NOT
——— c—— — i ———— ———— S—————— —————— ———— A———— ————p——— ——— —————h ———————— — —— —— —
9S
L
N
>
1027
BALDUS - DIRECT
TELLING ME THE REASON WHY YOUYRE WORRIED. BUT I DONT
UNDERSTAND THE PROBLEM.
MR. FORD: IM NOT, I“M HONESTLY NOT TRYING TQ DO, I
CAN TELL THE COURT FRANKLY, I, MY MAIN CONCERN RIGHT NOW. I
THINK THE COURT'S HEARD PROFESSOR BALDUS‘ ANALYSIS. I'M NOT
CONCERNED IN ANY OF THESE REGARDS AT THIS POINT ABOUT HAVING,
ABOUT THAT QUESTION. I BELIEVE THAT THAT’S BEFORE THE COURT.
MY CONCERN IS THAT IN THE APPELLATE COURTS TO WHICH
THIS CASE 1S LIKELY TO GO WHATEVER HAPPENS, I THINK THE COURT
HAS ALREADY ACKNOWLEDGED, I HOPE THE RECORD WILL REFLECT THAT
1 THE COURT HAS A SUBSTANTIVE BASIS FOR FINDING WHATEVER IT FINDS
WITH REGARD TO THE REALITY OF THE STATE OF CEORGIA DURING THIS
FERIOD.
THE COURT: THAT’S ASSUMING THAT THE REALILTY OF THE
STATE OF GEORGIA IS RELEVANT.
BUT ASSUMING THAT, MR, BOGER HAS BEEN PRESENT AND YOU
HAVE NQT BEEN PRESENT, SO YOU'RE NOT AS FAMILIAR WITH THE
RULINGS.
I THINK THE LINE I HAVE BASICALLY TRIED TO DRAW IS
WHERE SOME. SOME DEGREE OF SOPHISTICATION BEYOND JUST APPLYING
PERCENTAGES TO RAW NUMBERS 1S USED, SOME ADJUSTMENT BASED ON
SOME PERCEPTIONS OF WHAT GOES ON IN THE CRIMINAL JUSTICE SYSTEM
BEYOND JUST PERSONS CHARGED AND THINGS FLOW FROM THAT. WHERE ANY
OF THAT. WHERE ANY DEGREE OF SOPHISTICATION HAS BEEN INVOLVED IN
TRYING TO ADJUST FOR BACKGROUND FACTORS. I BELIEVE 1 HAVE
—
1023
BALDUS - DIRECT
po
t ADMITTED EVERY ONE OF THOSE DOCUMENTS. IF I HAVE NOT, I WILL BE
GLAD TO REVIEW ANY ALONG THOSE LINES THAT I HAVE NOT.
| WHAT THAT MEANS IS THAT THOSE DOCUMENTS WITH THAT DATA
IS IN EVIDENCE.
NOW, THAT DOESN‘T PRECLUDE PROFESSOR BALDUS FROM
RELYING ON WHATEVER IN THE WORLD HE WANTS TO RELY ON TO EXPRESS
HIS OPINION. HE CAN RELY ON ANYTHING HE WANTS TO RELY ON TO
EXPRESS HIS OPINION.
vv
©
«N
d
oo
OA
&
WO
N
BUT THOSE ARE JUDICATIVE FACTS WHICH ARE IN THE RECORD
—-
< CONCERNING WHATEVER DATA IS ON THOSE SHEETS, AND WHATEVER
Po
s
fd
s RESULTS COME FROM THE NUMBERS, WHICH MEANS THAT ANYBODY FROM
12 A JUDGE TO ANOTHER EXPERT CAN PLAY WITH THOSE NUMBERS AND DRAW
13 CONCLUSIONS. I THINK THAT'S ALL WE'RE DEALING WITH HERE.
14 MR. FORD: YOUR HONOR, --
13 THE COURT: IM NOT SURE THAT YOU NEED ANY OF IT. TELL
14 ME WHY YOU NEED A SINGLE ONE OF THOSE DOCUMENTS. EXCEPT PERHAPS
17 TO MAKE THE READER OF THE RECORD SEE IT?
: 1a MR. FORD: BECAUSE IT SEEMS TO ME, YOUR HONOR, THAT WE
Eo. 19 HAVE TO PROVE AT SOME LEVEL OF RELIABILITY THAT CERTAIN THINGS
20 OCCURRED IN THIS STATE DURING THIS PERIOD, AND IN THIS CASE.
3 AND THOSE DATA ARE THE QUESTIONS I HAVE, AT THIS POINT --— AND
22 I‘M ADVISED BY MY CO-COUNSEL THAT I“M PERHAPS PRESSING TCO HARD.
23 BUT ILL BE FRANK WITH THE COURT AND I“M SENSITIVE TO THAT, -—
29 THE COURT: I APPRECIATE YOUR CANDOR WITH THE COURT.
23 MR. FORD: LET ME VOICE ONE MORE CONCERN WITH THE
BALDUS - DIRECT
1 COURT.
¥ I WOULD, COULD THE COURT AT LEAST REQUIRE THE STATE
3 MAKE ANY EVIDENTIARY OBJECTIONS THAT IT HAS TO THE SUBSTANTIVE
4 FACTUAL BASIS AND THE SUBSTANTIVE FACTS THAT ARE TESTIFIED TO S50
5 THAT WE DON’T HAVE A SITUATION WHERE HEARSAY. THOSE KINDS OF
6 THINGS THAT HAVE NOT BEEN MENTIONED HERE. ARE BROUGHT UP ON
7 APPEAL AND .THEY’VE NEVER BEEN MENTIONED HERE AND WHEN WE CAN
3 DEAL WITH THEM HERE, IF THEY’RE REAL PROBLEMS, WHICH WE DON‘T
9 BELIEVE THEY ARE. WE BELIEVE WE‘VE GOT THERE.
10 BUT I DON’T WANT TO FIND OUT ON APPEAL THAT THE STATE
11 TAKES THE POSITION THAT SUPPORTING YOUR HONOR’S RULING BASED ON 12 ANOTHER GROUND, THAT THERE WAS SOME FAILURE OF FOUNDATION THAT
13 (WE WERE NEVER TOLD ABOUT, AND WHEN I FRANKLY HAVE NOT HEARD THE
14 STATE MAKE AN EVIDENCE OBJECTION TO THE VAST MAJORITY OF THIS,
15 THAT 1 UNDERSTAND TO BE AN OBJECTION UNDER THE RULE OF EVIDENCE.
16 1 BELIEVE THE STATE'S MADE ——
17 THE COURT: YOU MAY NOT HAVE BEEN IN THE COURTROOM --
18 MR. FORD: WELL, =--
- 19 THE COURT: -— WHEN MS. WESTMORELAND, WHEN IN DOUBT,
20 WHATEVER. HAS BECOME HER CONTINUING OBJECTION, AND IF YOU VE
|
21 |NEVER HEARD IT, YOU PROBABLY DON’T KNOW WHAT SHE’S BEEN SAYING
22 [WHEN SHE’S STOOD UP AND OBJECTED WHICH SHE HAS DONE WITH SOME
23 |REGULARITY.
|
24 THAT IS THE MISCHIEF IN LETTING LAWYERS COME AND GO.
25 SOME OF MY BRETHREN WOULD HAVE FUSSED WITH YOU AND MR. BOGER FOR
— — ——— — —— —— —— - ——— ——— ———————. ——— ————— ———. ——— —— J — A — ——— AAS tte Aan AA ———-.
1030
BALDUS - DIRECT
LEAVING AND LEAVING WITHOUT LEAVE OF THE COURT. I DON’T FUSS,
IVE TRIED LAWSUITS AND I“VE DONE EXACTLY WHAT YOU‘VE DONE. BUT
|
| THE MISCHIEF IS THAT YOU DON’T HAVE THE FACTUAL PREDICATE ON
WHICH TO ARGUE THE LAW THAT YOU KNOW, AND MR. BOGER DOESN’T HAVE
THE LEGAL PREDICATE ON WHICH TO ARGUE THE FACTS THAT HE KNOWS.
AND THAT PUTS YOU AT SOMEWHAT A DISADVANTAGE.
IT IS MY INTENT, SHOULD BE CLEAR FROM MY CONVERSATIONS
WITH MR. BOGER, THAT IF THIS CASE FALLS, IT NOT FALL ON SOME
SORT OF NIT-PICKY TECHNICAL GROUND. |
I HAVE INDICATED TQ YOU AND TO MR. BOGER THROUGHOUT
THAT I DO NOT WISH TO BE TROUBLED WITH UNSOPHISTICATED ACADEMIC
OPINIONS A LA WHAT I HEARD BEFORE. YOU HAVE BROUGHT ME
SOMETHING MUCH MORE SUBSTANTIAL THAN THAT. I HAVE TRIED TO LET
IN THAT MORE SUBSTANTIAL ANALYSIS SO THAT IT IS IN THE RECORD.
I DON’T KNOW WHAT ELSE YOU WANT IN THE RECORD AND IF
YOU WILL TELL ME WHAT YOU WANT IN THE RECORD, I WILL BE GLAD TO
CONSIDER IT ISSUE BY ISSUE. BUT I AM NOT GOING TO LET YOU PUT
IN MR. FORD’S COUNTERPART TO A SAFETY NET. OKAY?
MR. FORD: THANK YOU, YOUR HONOR.
THE COURT: ALL RIGHT. LET“S ADJOURN.
ARE YOU GOING TQ BE READY?
YOU“RE THROUGH. RIGHT?
MR. FORD: YOUR HONOR, --
MR. BOGER: I HAVE PERHAPS A HALF HOUR MORE. YOUR
HONOR.
)
BALDUS -
THE
MR.
THE
MR.
THE
MS.
THE
WESTMORELAND
DIRECT
COURT:
BOGER:
COURT:
BOGER?
COURT:
1031
HOW MUCH?
ID SAY HALF HOUR TO FORTY-FIVE MINUTES.
WELL, ILL LET YOU DO IT MONDAY MORNING.
ALL RIGHT.
AND THEN YOU/RE GOING TQ CROSS-EXAMINE?
WESTMORELAND: YES, YOUR HONOR.
COURT: ALL RIGHT. LET ME SEE MR. BOGER AND MS.
IN CHAMBERS, OFF THE RECORD.
WE-LL BE IN RECESS.
(COURT ADJOURNED FOR THE DAY.)
8
9
10
“w
w
0&
0
UO
+»
OQ
1032
BALDUS - DIRECT
C~E~R=T=1~F=1-C-A-T-E
UNITED STATES OF AMERICA
NORTHERN DISTRICT OF GEORGIA
I, JIM PUGH, OFFICIAL COURT REFORTER OF THE UNITED
STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF GEORGIA. DO
HEREBY CERTIFY THAT THE FOREGOING [ PAGES CONSTITUTE A TRUE
TRANSCRIPT OF PROCEEDINGS HAD BEFORE THE SAID COURT HELD IN THE
CITY OF ATLANTA, GEORGIA. IN THE MATTER THEREIN STATED.
IN TESTIMONY WHEREOF 1 HAVE HEREUNTO SET MY
HAND ON THIS [ DAY OF €
JIM PUGH
OFFICIAL COURT REPORTER
NORTHERN DISTRICT OF CEORGIA