IN THE SUPREME COURT OF TENNESSEE
AT NASHVILLE
February 4, 2003 Session
STATE OF TENNESSEE v. JERRY RAY DAVIDSON
Automatic Appeal from the Court of Criminal Appeals
Criminal Court for Dickson County
No. CR2232 Allen Wallace, Judge
No. M1998-00105-SC-DDT-DD - Filed October 20, 2003
ADOLPHO A. BIRCH, JR., J., dissenting.
I write separately to dissent from the majority’s holding that the evidence in this case is
sufficient to establish premeditation and to express my grave concern that the majority continues to
employ a proportionality review that I view as wholly inadequate. I join, however, Justice
Anderson’s dissent regarding the admission of Darla Harvey’s testimony.
A. Sufficiency of the Evidence
The majority concedes that the paucity of the evidence supporting premeditation presents the
Court with a “close” question as to whether premeditation was established beyond a reasonable
doubt; it then concludes that the “facts and circumstances as a whole” show premeditation. In my
view, however, the proof of premeditation is woefully lacking. The proof as I view it not only fails
to show premeditation, but demonstrates instead that the homicide resulted from impulsive behavior.
There was not even one scintilla of evidence in the record that the defendant engaged in any
preparation or planning for this crime. The defendant’s lack of preparation is evidenced in the
circumstances that brought the defendant and the victim together. On the evening of the crime, the
defendant and the victim left a bar together because the victim had no transportation, a circumstance
that the defendant could not have anticipated. Further, the defendant showed no concern for keeping
his conversation with the victim private, and he offered her transportation while in earshot of the
bartender.
The majority points to a “veritable arsenal of weapons” found in the defendant’s truck as
evidence of premeditation. The items found in the defendant’s truck, however, allow for numerous
other conclusions. To follow the majority’s reasoning, the defendant, who had already planned to
kill, had only to choose a victim and then from his “arsenal” find the means with which to
accomplish the crime. Other and more plausible conclusions emerge. The items in the defendant’s
truck may also suggest a rather nomadic person whose possessions are never far from hand or a
person well-equipped for any eventuality that may arise in a rural area. Just as clearly, it may be that
the defendant is simply an untidy person.
Further evidence points to the conclusion that the defendant did not act with premeditation.
The victim voluntarily left the bar with the defendant, and while the events that followed are unclear,
the evidence showed that the victim’s house was unlocked and that there was no sign of struggle or
blood in her home. The evidence did show that the victim’s blood was in the defendant’s truck, that
articles belonging to the victim were found in a field near her home, and that the defendant made a
statement to police that the victim might be found chained to a tree with her head and hands missing.
From this evidence, the majority concludes that the victim was first taken to a field near her home
and attacked in some way, and then she was taken to a secluded area, chained to a tree, and murdered
by the defendant, who was at all times acting with premeditation in committing this crime.
It is with the majority’s interpretation of the evidence that I cannot agree. Had there been
blood or signs of struggle in the field, or had the victim been found chained to a tree, this theory
would merit further consideration. However, such is not the case, and the evidence lends itself to
other more logical interpretations. One interpretation might be that the defendant, as a result of some
unexpected event, killed the victim, and after figuring out how to bury her, dumped her possessions
in the field. An interpretation such as this does not involve premeditation. Accordingly, the
evidence certainly does not prove beyond a reasonable doubt that the defendant committed a
premeditated murder, especially when the evidence of premeditation is entirely circumstantial.1
The defendant’s behavior after the crime also supports an interpretation that he did not plan
this murder. After the crime, the defendant suddenly and unexpectedly disappeared from his job and
his home, which resulted in a missing person report being filed. Again, this conduct is more
suggestive of a panicked response to an unexpected event rather than of a premeditated murder.
Finally, I disagree with the majority’s conclusion that the “nature of the mutilation” shows
a pre-existing intent to kill “carried out with care and precision.” The reason for the mutilation is
unascertainable; nothing, however, in this conduct reflects a calm or reflective mental state–either
before or after the murder. If anything, this conduct suggests extreme irrationality.
1
See State v. Crawford, 470 S.W.2d 610, 613 (Tenn. 1971) (holding that “[i]n order to convict on circumstantial
evidence alone, the facts and circumstances must be so closely interwoven and connected that the finger of guilt is
pointed unerringly at the de fendant and the defendant alone”); Smith v. State, 327 S.W .2d 308, 317 (Tenn. 1959 ) (“In
the effort to guard against improper verdicts, it is commonly stated that in determining the sufficiency of circumstantial
evidence, (1) all the essential facts must be consistent with the hypothesis of guilt, as that is to be compared with all the
facts proved; (2) the facts m ust exclude ev ery othe r reaso nable theory or hyp othesis except that of guilt; and (3) the facts
must establish such a certainty of guilt of the accused as to convince the mind beyond a reasonable doubt that the accused
is the one who committed the offense.”) (quoting Wharton's Criminal Evidence, 12th Edition, Vol. 3, Sec. 980, page
473); Pruitt v. State, 460 S.W.2d 385, 390 (Tenn. Crim. App. 1970) (stating that “to warrant a criminal conviction upon
circumstantial evidence alone, the evidence m ust be not only consistent with the guilt of the accused but it must also be
inconsistent with his innocence and must exclude every other reasonable theory or hyp othesis except that of guilt, and
it must establish such a certainty of guilt of the accused as to convince the mind beyond a reasonable doubt that he is the
one who com mitted the crime”).
-2-
Therefore, since the circumstantial evidence considered as a whole is just as easily construed
to support a less culpable mental state, I would find that a reasonable doubt as to the element of
premeditation remains.
B. Comparative Proportionality Review
Furthermore, I continue to adhere to the views expressed in a long line of dissents beginning
with State v. Chalmers, 28 S.W.3d 913, 920-25 (Tenn. 2000) (Birch, Jr., J., concurring and
dissenting), and elaborated upon in State v. Godsey, 60 S.W.3d 759, 793-800 (Tenn. 2001) (Birch,
Jr., J., concurring and dissenting), that the comparative proportionality review protocol currently
embraced by the majority is an inadequate protection from the arbitrary and disproportionate
imposition of the death penalty. See Tenn. Code Ann. § 39-13-206 (1997).
In addition to the substantive shortcomings of the procedure this Court has adopted, the
continuing failure of our database compilation system to generate a reliable source of information
for purposes of proportionality review further compromises reliable conclusions as to
proportionality. Pursuant to Rule 12 of the Tennessee Rules of the Supreme Court, a report is
required to be completed in all cases in which the defendant is convicted of first degree murder,
including cases in which the conviction is by a plea of guilty. Tenn. R. Sup. Ct. 12(1). There is a
form requiring specific information about the facts of the crime, the background of the defendant and
victim, and the sentence received. See id. The trial court is required to compile all the information
required and transmit the report to the Clerk of the Supreme Court within fifteen days after it has
ruled on the motion for new trial. Id.
The Rule 12 reports constitute the sole method by this Court of keeping a separate tally of
the sentences imposed in first degree murder convictions, and thereby, it is the best source of cases
to consider in a proportionality review. However, by just looking at my office files of the first degree
murder cases presented before this Court in the last six months of 2002, roughly twenty-nine percent
of first degree murders are not being included in these Rule 12 reports.2 Although the Department
of Corrections apparently keeps records from which all first degree murder convictions and sentences
could be extracted, as well as those convictions where the defendant was charged with first degree
murder but was convicted for a lesser offense, the Court relies on the Rule 12 reports that are filed
with the Clerk for its record of first degree murder convictions and sentences. This Court has been
greatly disappointed when revelations of omitted cases have shown the scarcity of these records.
Consequently, we have attempted to rectify the problems through both rule amendments and internal
procedures.
Despite published assurances from this Court that procedures to ensure the filing of Rule 12
reports are now in place, see Godsey, 60 S.W.3d at 785, it is apparent that the database compilation
process is still not working. Petitions for first tier review were filed for roughly forty-one first degree
2
My specific findings regarding the Rule 12 database are based on a review of the Rule 12 reports on file as
of January 1, 2003.
-3-
murder convictions; in twelve of these cases (involving thirteen defendants), no Rule 12 report was
filed.3 In one of the cases in which a Rule 12 report was filed, the report omitted the facts of the
murder.4 Such information is crucial in a proportionality review. The inevitable conclusion is that,
six years after our proportionality review process was carefully explained in State v. Bland, 958
S.W.2d 651, 661-74 (Tenn. 1997), and one year after remaining problems with reporting were
recognized in the dissent in Godsey, 60 S.W.3d at 793-800, twenty-nine percent of first degree
murder defendants (in my review, none of whom were sentenced to death5) are still not being
represented in the Rule 12 reporting system. The Rule 12 reporting system, as a method of enabling
this Court to practically consider similar cases in its proportionality review, has failed.
The omission of cases in which the death penalty was not imposed affects the integrity of our
analysis. How can our Court or the Court of Criminal Appeals determine whether the sentence of
death is proportional when many of the cases in which death was not imposed are not accessible for
review? In looking for cases similar in circumstances to the defendant’s, only death cases were
apparent from our initial search of the filed Rule 12 reports, a misleading result considering the
number of cases involving kidnapping and killing in which a life sentence was imposed.
For example, in the case of State v. Antonio Dewayne Carpenter, No. W2001-00580-CCA-
R3-CD, 2002 WL 1482799, at *1 (Tenn. Crim. App. Feb. 12, 2002), the defendant was convicted
of felony murder, premeditated murder, especially aggravated kidnapping, and especially aggravated
robbery. Id. According to the evidence presented at trial, the victim was abducted from a Sonic
restaurant at gunpoint. Id. at *3. She was struck in the head, and run over with a car several times.
Id. Her body was moved to a ditch and covered. Id. There was medical testimony that the victim
was alive when she was struck in the head, when her larynx was crushed, and when she sustained
the crushing injuries to her ribs, pelvis, abdomen, and chest. Id. at *4. Though the death penalty was
sought, the Fayette County jury sentenced the defendant to life without the possibility of parole. Id.
at *1. No Rule 12 report was filed in that case. The two co-defendants in that case were not eligible
for the death penalty (and thereby not part of the proportionality pool) because of mental retardation
and age. Id. at *1, n.2; See State v. Robert Lewis Carpenter, Jr., 69 S.W.3d 568 (Tenn. Crim. App.
3
State v. Lanard Keith Armstrong, M2000-02575-SC-R11 -CD (perm . app. denied); State v. Amos Brown,
E2000-00285-SC-R11-CD (perm . app. denied); State v. Jerry Baxter Graves, E2001 -001 23-S C-R11-C D (p erm. app.
granted); State v. Roger D ale H arris, E1992-0014-SC-R11-PC (perm . app. denied); State v. Mario Hawkins, M2000-
02901-SC-R11-CD (perm . app. denied); State v. Lavaya Demond Lee, E2001-00053-SC-R11-CD (perm. app. denied);
State v. Asata Lowe, E2000-01591-SC-R11 -CD (perm . app. denied); State v. Johnny M offitt, W2001-00781-SC-R11-CD
(perm. app. denied); State v. Clifford Peele, E1999 -009 07-S C-R11-C D (p erm app. d enied ); State v. Marthias S. Phillips,
M2000-02575-SC-R11-CD (perm . app. denied); State v. Fredrick Devill Rice, E2000-02389-SC-R11-CD (perm. app.
denied); State v. Kardius W ilks, W2001-02172-SC-R11-CD (perm. app. denied); and State v. Walter Wilson, W2001-
014 63-S C-R11-C D (p erm. app. d enied ).
4
State v. Alfonzo Williams, W2 001-004 52-SC-R11 -CD (perm. app . denied).
5
I have noted previously the disturbing conclusion that life cases are more likely to be o mitted from the d atabase
than death cases, the reby p lacing d efendants at a significant disadvantage in their efforts to locate those case s in which
claims of disproportionality may be based.
-4-
2001); State v. Glover, No. W2000-01278-CCA-R3-CD, 2001 WL 1078279 (Tenn. Crim. App. Sep.
14, 2001). The co-defendants, however, are included in the Rule 12 reporting system.
Antonio Dewayne Carpenter is a case involving kidnapping and mutilation that should be
included in the pool when determining the proportionality of the defendant’s sentence, but because
it was not properly reported, it was not considered by the Court of Criminal Appeals, nor is it
mentioned by the majority. Who knows how may other similar kidnapping cases are not included
in the Rule 12 reports. Even for the Rule 12 reports that are filed, delays and omissions compromise
the integrity of the resulting proportionality pool.
After Rule 12 reports are filed, they are compiled on a CD-ROM disk which attorneys may
obtain from the Administrative Office of the Courts. The specific factors of each case relevant to
proportionality review are supposed to be entered using “field codes.”6 Thereby, a death penalty
computer database is created which may be used by this Court and accessible to the litigants to find
similar cases. Additionally, an exact image of the Rule 12 report is scanned onto the disk.
The CD-ROM became available to attorneys in June 1999. Unfortunately, attorneys have
had mixed success in their ability to access the information on the disk from their computers. Aside
from these purely technical difficulties, there are also problems with incomplete entries. As stated
in one newspaper article, “hundreds of cases included in the database . . . are missing important
details about the crime, defendant, and victim.” Godsey, 60 S.W.3d at 796 (quoting John Shiffman,
Missing Files Raise Doubts About Death Sentences, The Tennessean (Nashville), July 22, 2001, at
A1). Indeed, this was the case in State v. Alfonzo Williams, W2001-00452-SC-R11-CD (perm. app.
denied), in which the Rule 12 report omitted the facts of the murder. Aside from determining from
the field codes that the case involved a robbery and killing by shooting, the entry on the database
does not explain the facts. Because Williams was appealed, the facts can be found by further
research of the Court of Criminal Appeal’s opinion. See State v. Williams, No. W2001-00452-
CCA-R3-CD, 2002 WL 1482695 (Tenn. Crim. App. Mar. 15, 2002).
Other types of omissions, even in appealed cases, are harder to track. For instance, though
there are field codes which indicate whether the death penalty was sought, the information on the
CD-ROM sometimes omits this information. In State v. Corley, No. 87-286-III, 1989 WL 41579,
at *1 (Tenn. Crim. App. Apr. 28, 1989), the unpublished Court of Criminal Appeals opinion states
that “[t]he record describes in graphic detail 15 hours of unrestrained criminal indulgence during
which one victim, an elderly man, was cheated, robbed, kidnapped and held captive and another
victim, a young woman, was murdered almost casually.” The defendant was sentenced to life. The
field codes entered for this defendant omit any reference to whether the death penalty was sought;
therefore, it is unclear whether the case should be included in the pool or not. If one used the field
codes to research only capital cases involving a kidnapping on the CD-ROM, this case would not
6
Each code represents a certain factor. For instance, the code “PHA4" should be entered for a case in which
the killing occurred during a rape, whereas the code “PHA8" should be entered if the killing occurred during a
kidnapping. Both should be entered if there was a kidnapping and a rape.
-5-
be shown. Nor would more traditional research on Westlaw or Lexis rectify the problem because
the opinion does not refer to the original sentence sought. See generally id. In other words, a
computer search for the word “death” or “capital” in the opinion will not reveal this case.
In general, the use of traditional methods of research is inadequate to fill the gaps in Rule 12
reporting. Such research is difficult, time consuming, and expensive. Furthermore, such research
necessarily reveals only those cases with published or unpublished written opinions, or in other
words, those cases which were appealed. Therefore, where a conviction is not appealed (a far more
likely occurrence where the sentence is life or life without parole), a Rule 12 report is the only
practically available source of information. Additionally, those first degree murder cases which are
appealed do not always state whether the death penalty was sought, a determination which presently
must be made in deciding whether to include a case in the pool.
Aside from the failure to enter certain information, the sheer delay in disseminating the
information affects the pool. In looking for cases with facts similar to the case pending before us,
one recent case that indicates current juries would impose a sentence of life without parole is State
v. Geraldrick Jones, W2002-00747-CCA-R3-CD (pending in C.C.A.). In Jones, after a first date,
the defendant brought the willing victim home with him; after an argument, he hit her several times
with his hands and with a five pound barbell weight. The victim attempted to escape, but the
defendant choked her, obtained a knife from the kitchen, and cut the victim’s throat seven times,
almost severing her head. He attempted to sever her limbs so that she would fit into the garbage
container. A Shelby County jury sentenced the defendant to life without parole. However, when
searching for cases similar to Davidson using the Rule 12 CD-ROM on January 8, 2003, this case
could not be found even though the Rule 12 report had been filed with the clerk eight months
earlier.7 It was not until an updated CD-ROM was distributed in mid-January of 2003, after the
publication of the Court of Criminal Appeals opinion, that the Jones case appeared on the CD-ROM
disk. Although the Administrative Office of the Courts has actively sought to update the disk every
six months,8 given that recent cases are more indicative of how recent crimes under recent
circumstances influence sentencing, every six months is not enough. Any system that denies access
to current sentencing in first degree murder cases denies a true proportionality review.
These flaws must be rectified before a death sentence may be affirmed. The problems with
the Rule 12 proportionality review system do not exist because of intentional or insincere efforts on
the part of our judicial system. The problem is that the goal–a system of fairly comparing cases so
that defendants are guaranteed that the termination of life is not a disproportionate
punishment–requires precise compliance with the adopted method of compiling and disseminating
pertinent information. The diversity of human nature exhibited in each particular county, jury, and
juror, inevitably leads to disparate sentences in capital cases. There is no other logical explanation
7
The Rule 12 report was filed with the Clerk on April 22, 2002.
8
W e are in no way criticizing the Administrative Office of the Courts. The manner in which they have
approached the formidable task of updating the CD-RO M d isks has be en imp ressive. Add itionally, their staff is always
helpful to our efforts to obtain more propo rtionality information.
-6-
for life sentencing in cases involving the brutal kidnapping, torture and murder of more than one
victim, see e.g., State v. Tatrow, No. 03C01-9707-CR-00299, 1998 WL 761829 (Tenn. Crim. App.
Nov. 2, 1998),9 and a death sentence in the kidnapping and murder in the case before us now. A true
proportionality review is critical to rectifying and balancing such disparate sentences. Without a
proportionality review that is based on all of the pertinent cases, this Court, as well as the Court of
Criminal Appeals, continues to risk affirming disparate sentences.
The problem of implementing an effective proportionality review is not unique to Tennessee.
Relying on findings from the Bureau of Justice Statistics and other legal studies, legal commentators
opine that the drop in the percentage of Americans who favor the death penalty (from eighty percent
to seventy percent),10 the drop in the number of people entering death row (which also declined
relative to the homicide rate),11 and the drop in the number of people executed,12 reflect the public’s
and the judicial system’s serious concerns about the administration of capital punishment. Adam
Liptak, Number of Inmates on Death Row Declines as Challenges to Justice System Rise, N.Y.
Times, Jan. 11, 2003, at A13. News articles show that in public opinion polls, forty to fifty percent
of the population believe the death penalty is not administered fairly, a perception that “has more to
do with practical problems [in fairly choosing those to be executed] than bedrock beliefs [opposing
capital punishment].” Id. The result is that although only twelve states plus the District of Columbia
have formally rejected execution as a form of punishment. Tracy L. Snell & Laura M. Maruschak,
U.S. Dep’t of Justice, Capital Punishment 2001 1 (Dec. 2002). In practicality, it is the minority of
locations that utilize the death sentence: “[Eighty] percent of American counties had no death
penalty convictions from 1983 to 1997, and another 10 percent had only one . . . .” Liptak, supra,
at A13. Such limited use of capital punishment not only indicates concerns with the system by those
responsible for imposing the law, but it also somewhat explains the inconsistency between sentences
in cases in which the facts are equally horrendous.
No sentence of death should be affirmed until the evident inconsistencies in the imposition
of sentences among juries can be checked through an adequate proportionality review. Both the
imposition and execution of death sentences should be recognized by this Court as too unpredictable
to meet the requirements of justice.
9
In Tatrow, over a two day period, the defendant kidnapped, tortured, and finally, killed two victims. 1998 WL
761829, at **1-6. T he Cumb erland County jury d eclined to impose the death pena lty. Id. at *1. Other cases involving
more than one victim, thereby being clearly more egregious that cases involving one victim, include State v. Lowe, No.
E2000-01591-CCA-R3-CD, 2002 WL 31 051631 (Tenn. Crim. App. Sept. 16, 2002), in which the defendant was
sentenced to life without the possibility of parole. No Rule 12 report has been filed in Lowe and it is impossible to tell
from the Co urt of Criminal App eals op inion whether the death penalty was sought.
10
Richard W illing, Death Penalty Gains Unlikely Defenders, USA Today, Jan. 7, 2003, 1A.
11
In 2001, 155 persons entered death row, the smallest number since 197 3. Ad am Liptak, Number of Inmates
on D eath Row Dec lines as Challenges to S ystem Rise, N.Y. Times, Jan. 11, 2003, at A13.
12
Indeed, most death sentenc es are never carried out. “Of the 779 people sen tenced to d eath in C alifornia in
the past four decades, for instance, 10 have been executed.” Id.
-7-
C. Conclusion
In conclusion, I find that the evidence is not sufficient to establish a premeditated killing
beyond a reasonable doubt, and that a death penalty review under the comparative proportionality
protocol presently used by the majority must not be upheld. Accordingly, I respectfully dissent.
_______________________________
ADOLPHO A. BIRCH, JR., JUSTICE
-8-