THORNBERRY, Circuit Judge:
Cleveland Colson was indicted by the Fulton County, Georgia, Grand Jury on February 15,1963, along with three other defendants, for robbery by use of force and arms. On April 1, 1963, Colson, represented by court-appointed counsel, pleaded guilty and was sentenced by the Fulton County Superior Court to a term of fifteen to twenty years. In February 1968 Colson filed a petition for writ of habeas corpus in federal district court. His action was stayed several times pending exhaustion of available remedies in the Georgia state courts. Finally, in September 1969, the district court, having concluded that petitioner had substantially exhausted his state remedies,
conducted an evidentiary hearing on petitioner’s contentions. At the hearing, petitioner presented evidence on the issues of systematic exclusion of Negroes from the Fulton County Grand Jury and ineffective assistance of counsel.
On October 14, 1969, the district court determined that a prima facie case of purposeful grand jury discrimination had been made out, and without reaching the adequacy of counsel question, ordered petitioner remanded to the custody of the State for speedy reindictment and trial. The State appealed, contending first that petitioner by pleading guilty had waived his right to challenge the construction of the grand jury, and also contesting the district court’s findings on the issue of grand jury composition. On June 9,
1970, this Court remanded the case to the district court for “findings of fact and conclusions of law regarding petitioner’s claim of ineffective counsel and the concomitant issue of the voluntariness of the plea of guilty.” Colson v. Smith, 5th Cir. 1970, 427 F.2d 143. Since the district court had already conducted a full hearing on the issue of counsel’s competency, all that remained to comply with our order
of
June 9 was to enter findings of fact and conclusions of law based on the evidence previously presented. Accordingly, the district court entered a final order dated July 17,1970, concluding that petitioner’s plea of guilty was the product of ignorance, fear, and the ineffective assistance of counsel, and again ordering petitioner’s release subject to the State’s right to reindict him. The State returns to this Court now, contesting not only the district court’s findings in its order of October 14, 1969, but also the latest findings of the district court on the issue of effectiveness of counsel.
I.
At the outset we advert to the settled rule in this Circuit that a voluntary plea of guilty waives all non jurisdictional defects, including the right to challenge the construction of the grand jury. Williams v. Smith, 5th Cir. 1970, 434 F.2d 592; Throgmartin v. United States, 5th Cir. 1970, 424 F.2d 630. Under this rule, were we to find that petitioner’s guilty plea was voluntarily entered, we would be precluded from any consideration of the issue of grand jury composition. Thus we must dispose first of petitioner’s attack on his plea of guilty.
It is clear that a defendant is entitled to the effective assistance of counsel in determining how to plead and in making his plea, and can attack his conviction collaterally if he is not given this right. 1 C.A. Wright, Federal Practice and Procedure § 171 (1969); Davis v. United States, 5th Cir. 1967, 376 F.2d 535. Moreover, the Supreme Court has said that this threshold right to the assistance of counsel is no less momentous to an accused who must decide whether to plead guilty than to an accused who stands trial. Von Moltke v. Gillies, 332 U.S. 708, 721, 68 S.Ct. 316, 322, 92 L.Ed. 309 (1948). In defining the
scope of counsel’s duty to an accused client,
however, courts have distinguished between the defendant who is standing trial and the defendant who is pleading guilty. This Court recently stated that “the only required duty of counsel under the most liberal construction when a plea of guilty is entered is that counsel * * * should ascertain if the plea is entered voluntarily, and knowingly.” Lamb v. Beto, 5th Cir. 1970, 423 F.2d 85, 87. Similarly, the Court of Appeals for the District of Columbia, in an opinion by the now Chief Justice of the United States, Warren Burger, has stated that “ineffective assistance of counsel, as opposed to ignorance of the right to counsel, is immaterial in an attempt to impeach a plea of guilty, except perhaps to the extent that it bears on the issues of voluntariness and understanding.” Edwards v. United States, 1958, 103 U.S.App.D.C. 152, 256 F.2d 707, 709.
See also
Kress v. United States, 8th Cir. 1969, 411 F.2d 16; Alaway v. United States, D.C.Cal.1968, 280 F.Supp. 326. The reason for this narrow prescription of counsel’s duty to a client who pleads guilty was explained by the then Judge Burger in
Edwards, supra,
as follows:
It must be realized this is not a case in which proof of guilt depended upon a trial. In such cases, the accused usually relies to a great extent on counsel to conduct an effective defense, because the accused does not know enough of the law to do so himself. While the accused may have to take the consequences of a poor defense, he may at least say the fault was not his own. But this is not so when he pleads guilty. Here the deed is his own; here there are not the baffling complexities which require a lawyer for illumination; if voluntarily and understanding^ made, even a layman should expect a plea of guilty to
be treated as an honest confession of guilt and a waiver of all defenses known and unknown. And such is the law.
256 F.2d 707, 709.
We subscribe fully to the principle that guilty pleas are meant to be, and should be, final. And if there was ever any doubt in our minds of the inviolability of that principle, there is certainly no longer any doubt after the Supreme Court’s recent decisions declaring with unmistakable clarity its firm commitment to the finality of guilty pleas. McMann v. Richardson, 397 U.S. 759, 90 S.Ct. 1441, 25 L.Ed.2d 763 (1970); Parker v. North Carolina, 397 U.S. 790, 90 S.Ct. 1458, 25 L.Ed.2d 785 (1970); Brady v. United States, 397 U.S. 742, 90 S.Ct. 1463, 25 L.Ed.2d 747 (1970). A necessary result of the seriousness with which courts treat pleas of guilty, however, is that courts have a concomitant responsibility to assure that defendants who plead guilty do so voluntarily and knowingly.
Free access — add to your briefcase to read the full text and ask questions with AI
THORNBERRY, Circuit Judge:
Cleveland Colson was indicted by the Fulton County, Georgia, Grand Jury on February 15,1963, along with three other defendants, for robbery by use of force and arms. On April 1, 1963, Colson, represented by court-appointed counsel, pleaded guilty and was sentenced by the Fulton County Superior Court to a term of fifteen to twenty years. In February 1968 Colson filed a petition for writ of habeas corpus in federal district court. His action was stayed several times pending exhaustion of available remedies in the Georgia state courts. Finally, in September 1969, the district court, having concluded that petitioner had substantially exhausted his state remedies,
conducted an evidentiary hearing on petitioner’s contentions. At the hearing, petitioner presented evidence on the issues of systematic exclusion of Negroes from the Fulton County Grand Jury and ineffective assistance of counsel.
On October 14, 1969, the district court determined that a prima facie case of purposeful grand jury discrimination had been made out, and without reaching the adequacy of counsel question, ordered petitioner remanded to the custody of the State for speedy reindictment and trial. The State appealed, contending first that petitioner by pleading guilty had waived his right to challenge the construction of the grand jury, and also contesting the district court’s findings on the issue of grand jury composition. On June 9,
1970, this Court remanded the case to the district court for “findings of fact and conclusions of law regarding petitioner’s claim of ineffective counsel and the concomitant issue of the voluntariness of the plea of guilty.” Colson v. Smith, 5th Cir. 1970, 427 F.2d 143. Since the district court had already conducted a full hearing on the issue of counsel’s competency, all that remained to comply with our order
of
June 9 was to enter findings of fact and conclusions of law based on the evidence previously presented. Accordingly, the district court entered a final order dated July 17,1970, concluding that petitioner’s plea of guilty was the product of ignorance, fear, and the ineffective assistance of counsel, and again ordering petitioner’s release subject to the State’s right to reindict him. The State returns to this Court now, contesting not only the district court’s findings in its order of October 14, 1969, but also the latest findings of the district court on the issue of effectiveness of counsel.
I.
At the outset we advert to the settled rule in this Circuit that a voluntary plea of guilty waives all non jurisdictional defects, including the right to challenge the construction of the grand jury. Williams v. Smith, 5th Cir. 1970, 434 F.2d 592; Throgmartin v. United States, 5th Cir. 1970, 424 F.2d 630. Under this rule, were we to find that petitioner’s guilty plea was voluntarily entered, we would be precluded from any consideration of the issue of grand jury composition. Thus we must dispose first of petitioner’s attack on his plea of guilty.
It is clear that a defendant is entitled to the effective assistance of counsel in determining how to plead and in making his plea, and can attack his conviction collaterally if he is not given this right. 1 C.A. Wright, Federal Practice and Procedure § 171 (1969); Davis v. United States, 5th Cir. 1967, 376 F.2d 535. Moreover, the Supreme Court has said that this threshold right to the assistance of counsel is no less momentous to an accused who must decide whether to plead guilty than to an accused who stands trial. Von Moltke v. Gillies, 332 U.S. 708, 721, 68 S.Ct. 316, 322, 92 L.Ed. 309 (1948). In defining the
scope of counsel’s duty to an accused client,
however, courts have distinguished between the defendant who is standing trial and the defendant who is pleading guilty. This Court recently stated that “the only required duty of counsel under the most liberal construction when a plea of guilty is entered is that counsel * * * should ascertain if the plea is entered voluntarily, and knowingly.” Lamb v. Beto, 5th Cir. 1970, 423 F.2d 85, 87. Similarly, the Court of Appeals for the District of Columbia, in an opinion by the now Chief Justice of the United States, Warren Burger, has stated that “ineffective assistance of counsel, as opposed to ignorance of the right to counsel, is immaterial in an attempt to impeach a plea of guilty, except perhaps to the extent that it bears on the issues of voluntariness and understanding.” Edwards v. United States, 1958, 103 U.S.App.D.C. 152, 256 F.2d 707, 709.
See also
Kress v. United States, 8th Cir. 1969, 411 F.2d 16; Alaway v. United States, D.C.Cal.1968, 280 F.Supp. 326. The reason for this narrow prescription of counsel’s duty to a client who pleads guilty was explained by the then Judge Burger in
Edwards, supra,
as follows:
It must be realized this is not a case in which proof of guilt depended upon a trial. In such cases, the accused usually relies to a great extent on counsel to conduct an effective defense, because the accused does not know enough of the law to do so himself. While the accused may have to take the consequences of a poor defense, he may at least say the fault was not his own. But this is not so when he pleads guilty. Here the deed is his own; here there are not the baffling complexities which require a lawyer for illumination; if voluntarily and understanding^ made, even a layman should expect a plea of guilty to
be treated as an honest confession of guilt and a waiver of all defenses known and unknown. And such is the law.
256 F.2d 707, 709.
We subscribe fully to the principle that guilty pleas are meant to be, and should be, final. And if there was ever any doubt in our minds of the inviolability of that principle, there is certainly no longer any doubt after the Supreme Court’s recent decisions declaring with unmistakable clarity its firm commitment to the finality of guilty pleas. McMann v. Richardson, 397 U.S. 759, 90 S.Ct. 1441, 25 L.Ed.2d 763 (1970); Parker v. North Carolina, 397 U.S. 790, 90 S.Ct. 1458, 25 L.Ed.2d 785 (1970); Brady v. United States, 397 U.S. 742, 90 S.Ct. 1463, 25 L.Ed.2d 747 (1970). A necessary result of the seriousness with which courts treat pleas of guilty, however, is that courts have a concomitant responsibility to assure that defendants who plead guilty do so voluntarily and knowingly. One aspect of this responsibility is manifested in Rule 11 of the Federal Rules of Criminal Procedure, which imposes a mandatory duty on the trial court to ascertain that a guilty plea is made voluntarily and with understanding of the nature of the charge.
See also
McCarthy v. United States, 394 U.S. 459, 89 S.Ct. 1166, 22 L.Ed.2d 418 (1969). We think that an equally important aspect of the courts’ responsibilities in this problem area is the protection of the accused’s right to the benefit of reasonably competent counsel in making his plea,
especially when that plea is entered on advice of counsel. One cannot read the Supreme Court’s opinions in
McMann, Parker
and
Brady, swpra,
without being impressed by the significance the Court attached to the
role of counsel
in the process of deciding how to plead. In all three cases, it was obvious that the Supreme Court envisioned a system under which the defendant, advised by reasonably competent counsel, makes an informed and conscious choice.
See also
Yon Moltke v. Gillies,
supra.
When this system functions satisfactorily, it is both fair and reasonable to expect that a defendant who has made his choice and received whatever benefits flow therefrom be required to live by that choice. In any particular case in which the system fails, however, it is the courts’ duty to supply relief.
Bearing all the foregoing in mind, we turn now to the specific allegations and findings of the instant case.
There is testimony from petitioner in this case that his court-appointed counsel informed him that one of his co-defendants was going to testify against him, that counsel could not undertake to defend him in a jury trial unless petitioner could pay him, and that the best thing petitioner could do was plead guilty. Petitioner’s counsel contradicted much of this testimony, but it was uncontroverted that petitioner adhered to his decision to plead not guilty until the day the case was called. Furthermore, it was uncontradicted that there was considerable discussion between the petitioner and his counsel, on the one hand, and the state trial judge, on the other, before petitioner entered his final plea. During this colloquy, it appears that petitioner resisted pleading guilty for some time before finally re
lenting.
It also appears that although counsel knew beforehand that petitioner wanted to plead not guilty, counsel appeared in court on the day the case was called completely unprepared to go to trial. On this record the trial judge
made a fact finding that petitioner’s guilty plea was the product of ineffective assistance of counsel. This finding was essentially a credibility decision, which we, of course, will not disturb except on a showing of clear error. United States v. Saunders, 5th Cir. 1970, 435 F.2d 683; United States v. Schoen, 5th Cir. 1970, 434 F.2d 931; Bickers v. Cran-ford, 5th Cir. 1970, 433 F.2d 955. In this case, however, we go beyond the traditional principles of restraint in appellate review to uphold the district court. We agree further with the district court’s finding that under either version of the facts, the evidence in this case demonstrates that petitioner was ineffectively represented in his state court proceeding. We reach this conclusion even in light of the narrow duty imposed on counsel representing a defendant who pleads guilty.
See
Lamb v. Beto,
supra.
For we believe that the evidence in this case supports the conclusion that counsel actively tried to convince petitioner to plead guilty in the face of petitioner’s repeated protestations of innocence; and to offset this evidence, there was no evidence that counsel’s advice to plead guilty was based on any evaluation of petitioner’s chances had he gone to trial.
In other words, we do not have before us a case of “reasonably competent” advice which, with the benefit of hindsight, turned out to be mistaken.
Cf. McMann, Parker,
and
Brady, supra.
We are of the opinion that on this record there is no way that an appellate court could conclude that counsel had fulfilled his duty to ascertain that his accused client’s plea of guilty was voluntarily and understandingly made. We therefore affirm the district court’s order to set aside the conviction. Having thus invalidated the guilty plea, we proceed to petitioner’s challenge to the racial composition of the grand jury that indicted him.
II.
We are not unfamiliar with the Fulton County Grand Jury selection process. On more than one occasion in the past, both the Supreme Court and this Court have held that this system, which uses segregated tax rolls as the main source for the master grand jury list, as a matter of law presents an “opportunity for discrimination.”
Whitus v.
Georgia, 385 U.S. 545, 552, 87 S.Ct. 643, 647, 17 L.Ed.2d 599 (1967); Jones v. Smith, 5th Cir. 1969, 420 F.2d 774. In the instant ease, the grand jury that indicted petitioner in the first term of 1963 was selected from the 1959 master grand jury list, which was compiled under the aforementioned system in the following manner: Six white jury commissioners selected the names for the 1959 master grand jury list from the tax returns of 1959, which were made on separate slips of paper, each colored according to the race and residence of the taxpayer. White forms were used for white taxpayers living inside the City of Atlanta. Pink forms were used for Negro taxpayers living inside the city. Blue forms for white taxpayers living outside the city, and green forms for Negroes living outside the city. From among these slips, one Commissioner would choose a slip, call out the name, and if any of the Commissioners knew the named person, that Commissioner would state whether the person was "qualified.” The statutory standard used to determine whether a taxpayer qualified was that a person be an experienced, upright and intelligent citizen of the community.
If none of the Commissioners knew the person, then the Commissioners searched the old petit jury books for the name. If the person had served on the petit jury for several years, he qualified for the grand jury. Other indicia used by the Commissioners to determine intelligence, uprightness and experience were whether the taxpayer was a homeowner and had lived in the community for many years; or whether the taxpayer had been at his present job for a reasonable length of time. The names selected in the foregoing manner were then put on the master grand jury list, and on separate slips of paper, none of which were racially designated. These slips were placed in one compartment of the grand jury box. The jury venire for each term of court was then drawn at random from this box by a local judge. The slips drawn were then placed into another compartment of the grand jury box to separate them from the names yet uncalled.
When a state or county utilizes such a potentially discriminatory selection process, the only proof required to make out a prima facie case is proof of an impermissible disparity in racial composition between the tax rolls and the venire or list from which the
particular
grand jury is drawn.
Jones supra,
at 776. Petitioner presented evidence that in his case the Fulton County Grand Jury Selection Process produced the following statistical results. According to the 1960 census, the Fulton County population consisted of 69.3 white adult males and 36.7 Negro adult males. The
1959 master grand jury list was selected by an all-white jury commission and the grand jury that indicted petitioner was 100% white. The foregoing evidence was all direct evidence and was not objected to by the State. Of course, alone it would not be sufficient to make out .a prima facie case.
But petitioner went further and presented evidence of the racial composition of the tax rolls for 1959, the 1959 master grand jury list, and the jury venire from which petitioner’s grand jury was drawn, as well as evidence of the racial composition of other grand jury venires for the term of court preceding the term in which petitioner was indicted. On each of these issues, however, petitioner was unable, because of the alleged unavailability of certain statistical sources,
to present first-hand evidence of racial composition. Rather, petitioner resorted to the alternative methods of proof discussed below.
Petitioner did not show, for example, the racial breakdown of the Tax Rolls for 1959, which was the source of the 1959 master grand jury list, because of the alleged unavailability of racial information for that year.
Instead, petitioner relied on records from later years showing a constant figure of approximately 15 percent for Negro taxpayers in Fulton County. Neither did petitioner offer direct proof of the racial breakdown of the 1959 master grand jury list, from which the jury venires in the instant case were drawn. Failure of direct proof on this point was also caused by the alleged unavailability of racial information in the 1959 tax rolls. Since the racial identity of the persons listed on the 1959 master list could not be ascertained directly, petitioner checked the names and addresses on the 1959 list against the same names on the 1962 tax rolls. Petitioner did not, however, check the race of
every
individual on the 1959 master list. Instead, he offered in evidence a random sample of seventy-five names based on a selection of every fortieth name of the 3007 names on the master list. Thus, petitioner took each fortieth name and address on the 1959 master grand jury list, searched the 1962 tax rolls for the same name, and if the address was identical, assumed that it was the same person and relied on the racial designation of that person on the 1962 tax rolls for proof of that particular taxpayer’s race. If the address on the 1962 tax rolls was different from the address on the 1959 list, but there was only one such name on the 1962 tax rolls, petitioner assumed again that it was the same person and relied on the racial designation of the 1962 tax rolls. If there were two such names on the 1962 rolls, and the addresses were different, petitioner designated the race as “unknown.”
By referring to the 1962 tax rolls, petitioner also compiled statistics showing the racial breakdown of the grand jury venire from which the indicting grand jury was selected, as well as the racial breakdown of all the Fulton County grand jury venires and grand juries during 1962. These methods of proof
produced the following statistical profile of petitioner’s case:
PERCENTAGES
White Negro Unknown
Fulton County Population
(1960 census-males over 21) 69.3 30.7 0
Fulton County Tax Rolls 86.0 14.0 0
Jury Commissioners 100 0 0
Grand Jury that indicted petitioner 100 0 0
Grand Jury Venire from which indicting grand jury selected 90 0 10
Grand juries during six terms of 1962 and first term of 1963 98.8 1.2 0
Grand jury venires from which preceding selected 84.6 .7 14.7
Random Sample of 75 names, from 1959 Master Grand Jury List 90.7 1.3 9
The district court was satisfied that this proof made out a prima facie case of systematic exclusion of Negroes from the grand jury, and we agree. As the district court pointed out in its opinion, petitioner was able to discover the race of 90% of the panel from which his grand jury was selected, and no Negroes were included in that number; the race of 85% of the grand jury panels from which the first 1963 grand jury and all the 1962 grand juries were drawn and fewer than 1% were Negroes; and the race of 91% of a random sample of 75 names from the 1959 grand jury list, of which fewer than 2% were Negroes. Assuming that petitioner’s proof methods are acceptable, we believe that these statistics make out a prima facie case.
The State does not argue that petitioner’s statistics on their face fail to show an impermissible disparity between the racial composition of the tax rolls and the racial composition of the grand jury venire. It is petitioner’s proof methods that the State so vigorously contests. After reviewing the State’s arguments with great care,
we find that we are unpersuaded by them for two reasons. First, we note that in certain respects, petitioner went a good deal farther in proving his case than he was required to go under the standards enunciated in
Jones
and
Whitus, supra,
for cases in which segregated tax rolls have been used as the source for the master grand jury list. For example, petitioner offered proof of the racial composition of the 1959 master grand jury list. While proof of the percentage of Negroes and white persons in the pool of jurors would have been required in a case not involving the use of segregated tax rolls, Rabinowitz v. United States, 5th Cir. 1966, 366 F.2d 34; Roach v. Mauldin, 5th Cir. 1968, 391 F.2d 907, as we pointed out earlier in this opinion, that requirement was relaxed for cases of the type under review here,
Jones, supra.
Petitioner also showed the racial composition of the grand jury ve-nires other than the one from which his indicting grand jury was selected. This evidence, too, was more than
Whitus
and
Jones
require.
The second reason we are unpersuaded by the State’s arguments is that we believe that this case presents just the type of problem that the prima facie evidence device was intended to deal with. Proving that there has been racial discrimination in the selection of grand jurors is a very difficult task for parties challenging their indictments because of the large amount of statistical data that must be gathered. The task is much easier for States because this type of in
formation is peculiarly within the State’s knowledge. In such cases, it clearly would be unfair to indictees to insist on exacting methods of proof. McCormick on Evidence § 309 (1954). At the same time, however, the State must be protected from being required to produce statistics on grand jury composition every time someone raises an unsubstantiated allegation of systematic exclusion. Willis v. Smith, 5th Cir. 1970, 434 F.2d 1029. Herein lies the usefulness of the prima facie presumption in jury discrimination cases. On the one hand, it requires parties who challenge their indictments to carry the buden of proof up to a point. On the other hand, however, it protects those parties once they have raised sufficient doubts about the system, for it prevents States from smugly sitting back and watching indictees labor under an almost impossible burden of proof when the States could easily resolve any doubts or fill any gaps with information available at their fingertips.
Applied to the instant case, we think the prima facie presumption has served the Court and the parties well. Petitioner has been able to prove his case up to a point, but for lack of information (which was at one time, at least, in the hands of the State), he has been unable to go farther. We think that he has shown this Court enough to raise substantial doubts about the constitutionality of his indictment. Prima facie evidence is, after all, not conclusive unless nothing is done by the party on the other side to refute it. 9 Wigmore on Evidence § 2494 (1940). In this case, the State could easily have come forward and refuted or disproved petitioner’s case with facts of its own, if those facts had been in the State’s favor. Instead, however, the State chose to complain about petitioner’s proof methods. It is true that petitioner’s methods were oblique, but they were the best he could do, and the State has not demonstrated to us that they were unreliable. It may also be true that in the jury composition cases decided by this Court to date, there has always been first-hand knowledge of racial composition. But this does not mean that there is only one acceptable method of proving systematic exclusion.
While we, of course, do not fault the State for making its arguments with respect to petitioner’s proof methods, we decline to accept those arguments in a case of this type. For when a defendant’s right to be indicted by a grand jury from which members of his race have not been systematically excluded is at stake, justice will not permit us to indulge in games of courtroom strategy. In this case, petitioner presented enough evidence of a disparity between the tax rolls and the grand jury venire to justify requiring the State to come forward and explain. Since the State declines to explain, we affirm.