Circuit Judges, join, dissenting:
Respectfully, I dissent. If a judge punishes a criminal defendant because the defendant declines to plead guilty and, instead, avails himself or herself of the constitutionally guaranteed right to a trial, such action is unconstitutional. The fact that such punitive measure is presented in the context of plea bargaining does not make it lawful.
No guilty plea was entered in this case. Therefore, we are examining the reverse side of the coerced plea situation. It is not contended that petitioner was unlawfully coerced into pleading guilty. It is alleged that he was punished, by enhanced sentence, because he would not accommodate the criminal justice system. He insisted upon his constitutionally guaranteed right to trial. Two efforts had been made-one before trial and one during trial-to obtain a guilty plea in petitioner’s case. The judge participated in each of the discussions. The judge announced, on each occasion, that the sentence would be confinement for twenty years upon a conviction based upon a guilty plea. Instead, the conviction was upon a jury verdict. The same judge who had found and announced that a twenty year sentence would have been appropriate had petitioner pleaded guilty sentenced petitioner to thirty-three years confinement after trial.
Our majority flatly holds:
Even if the trial provided no additional evidence of character, the mere fact that Jimmy Frank refused to acknowledge his guilt and showed no willingness to assume responsibility for his conduct may have lead the judge to conclude that this defendant lacked potential for rehabilitation thus justifying the imposition of a greater sentence than that offered in exchange for a guilty plea.
(At 885). I view this holding as jurisprudence foreign to our own and contrary to our Constitution. I do not dispute that there are lands in whose courts it is necessary that an accused confess that of which he is charged in order to avoid inappropriately increased punishment for the additional antisocial act of putting the accuser to its proof. I do dispute that in our system the mere fact that an accused refuses to plead guilty to the state’s charge is a punishable offense.
We receive this case cloaked in the often controversial garb of what is known as “plea bargaining.” Of the many forms of that method of resolving disputes between the executive and defendants, the one here presented is “sentence bargaining.”1
Plea bargains based upon the sentence to be imposed provoke great controversy and may yet be the most helpful to the proper administration of justice. In most cases it is the sentence which concerns the defendant most. The resolution of the uncertain*894ty inherent in the offended statutes’ maximum allowable punishment into certainty as to an appropriate sentence for the particular defendant guilty of the specific offending acts may provide the basis for resolution of all other issues.
For me, I must indulge in some investigation of this practice-before I can appraise the particular happenings giving rise to the case before us. I embark upon that investigation because, as a distinguished jurist wrote some nine decades ago, “When the right point of view is discovered, the problem is more than half solved.”2
I. The Right Point of View
Hardly anyone who seriously contemplates the role of courts in dispute resolution would deny that settlement of disputes, freely made by well-informed, well-advised parties to the disputes is to be commended. Courts exist to resolve those disputes which cannot be appropriately resolved otherwise and the availability of courts is some assurance that neither party to a dispute need accept an inappropriate resolution.
In some cases, what has been agreed upon between parties cannot be effectuated until the court passes upon its appropriateness. The agreement reached by a guardian ad litem for a minor or otherwise incompetent ward has historically required review by the court before it binds the parties. Custody agreements between litigating parents of children of disrupted marriages, stockholder derivative suit settlements, some compromises by trustees in bankruptcy, and agreeable terminations of class actions require that a judicial officer investigate and affirm that the agreements are appropriate. Sentences to be imposed on defendants convicted upon pleas of guilty are, I submit, other such cases. The mere fact that court review in such eases is required does not mean that the settlements are discouraged. Agreeable resolutions of this sort serve the parties and serve the administration of justice. Often such resolutions conserve great additional investments of capital funds and expense disbursements from the public. They also save the disputing parties from incurring unnecessary and often great costs, inconveniences, and uncertainties. To suggest that settlements are somehow evil would be like positing that cures of physical problems through clinical therapy are to be avoided if surgery be available. Courtrooms must be available in courthouses, and surgical amphitheatres are a part of hospitals, but recourse should be had to neither if proper results can be obtained without their use.
I submit that the benefits of agreeable resolution of disputed issues extend to disputes between the executive and citizens charged with crime if agreements are reached through proper procedures. Further, criminal cases can be resolved through negotiation without involving threats by the court to punish those who choose trial.
In the settlement of civil cases and disputes referred to earlier, some practices are worth noting:
1. Negotiations are carried on between parties with adverse interests, one to the other.
2. While it cannot be denied that settlement of such cases is beneficial to the court as an institution and to its presiding judge, the time, efforts and facilities of which would otherwise be devoted to trials, the judge is not one of the adversaries. The judge is neutral, neither adverse to nor advocate for, either party.
3. During negotiations, but before resolution, demands and offers are proposed differing from the ultimate arrangement upon which agreement is reached.
4. Negotiations involve proposals coupled with threats. An offer to concede a position by one of the adversaries is necessarily coupled with a threat to return to that position should the concession not result in agreement.
*8955. Nevertheless, when the process has been concluded, the product of it is endorsed by both parties to the dispute as the appropriate result, all things being considered. Thus, the former disputants become mutual advocates of the settlement.
6. In those cases where court approval is required, the settlement is presented to the court, by both former disputants, as the appropriate resolution of all the issues given due consideration to all legitimate interests.
7. Where the advocates of those competing interests have performed properly, the settlement is appropriate and the court will so find by its approval. Where one has been lacking in attention, effort or skill, the inappropriate resolution reflecting such shortcomings will be recognized by the court’s rejection.
A civil case is commenced by the filing of a petition or a complaint setting out, among other things, an ad damnum. The demand is seldom inadequate! Realistically, it is not expected that ultimate recovery will be so great as the demand — though none can assure the defendant that it will not. Even in cases of apparent liability, cautious defendants will not permit the case to go into default. Defensive pleadings are filed, and preparations are made to urge whatever may properly be urged to defeat the claim.
When such a case is settled, the defendant, in effect, agrees to change its position from one of no liability to a confession of liability — but not liability for the ad damnum set out in the action. Instead the defendant withdraws defenses if the amount it pays be limited to the certain amount fixed by the agreement.
In pure theory, where there is no reasonable expectation of defeating the asserted liability, a defendant would do as well to default and appear only to contest damages. Prudence usually dictates, however, that it not open the floodgates until assurance has been obtained as to how much water will rush through!
When proceedings are commenced against a person charged with crime, the assertion is made that the defendant is guilty and the maximum punishment which may be exacted is set forth in the statute. In many cases, it is not realistically expected that, upon conviction by a jury verdict or a guilty plea, the maximum penalty would be imposed-though none can assure the defendant that it will not. Even in cases where guilt is apparent, cautious defendants often will not enter a guilty plea as their first response to the accusation or indictment.
Often, the prosecuting, attorney advocating the interests of the public and the defendant, personally or by counsel, advocating the particular interests of the defendant, can, within the maximum penalty provided by law, arrive at an appropriate resolution of the case. It seems proper and constructive in such cases for the defendant to be permitted to withdraw his defenses if the consequences be limited to those certain consequences fixed by the agreement. It remains for the court to pass upon the appropriateness of the agreed sentence, but the judge has the benefit of the product of the negotiations. If one advocate or the other has been less than able, the inappropriateness of the agreement can be detected and the agreement disapproved.
I emphasize that the purpose of the negotiations is to reach agreement upon an appropriate sentence. While it may properly be said that the process involves the offer of concessions to the defendant, the ultimate concession on the part of the prosecutor is to abandon the inherent uncertainty of advocating the maximum sentence permitted under the law and to join with the defendant in advocating a certain sentence, arrived at through negotiation, as the appropriate one in the particular case. The prosecutor has the right to advocate even an inappropriately harsh sentence. The concession is the giving up of that right. The prosecutor has no right to the imposition of the maximum sentence, so he cannot concede any such right in the bargaining process. The defendant has the absolute right to advocate his innocence. By agree*896ing to plead guilty, he concedes that right. If he be guilty, he has no right to a judgment of acquittal, and his plea agreement is not a concession of any such right. Properly conducted, plea bargaining involves the concession by each adversary of his right to advocate to uncertain result in exchange for certainty. That certainty ought to be as to an appropriate sentence. It ought not involve the offer of inappropriate leniency to a defendant, in derogation of the prosecutor’s duty to society, nor the threat of inappropriate punishment should the defendant decline to plead. The prosecutor, alone, cannot make such an inappropriate offer or threat. The prosecutor cannot control sentencing. The judge decides what the sentence shall be.
Central to our obtaining the right point of view of this case is that plea bargaining was not between the legitimate adversaries, the executive represented by the prosecutor and the accused represented by defense counsel. The judge, whose proper role demands that he be adverse to neither, entered into the negotiations. He was given a description of the facts and circumstances of the crime charged and of the defendant sufficient unto him for the determination of an appropriate sentence.3 It must have been sufficient for the judge proceeded, in his judicial capacity, to state that, upon conviction based on a plea of guilty, he would impose sentence of twenty years confinement.
Presumably, as judge of the court, he determined in this particular case that a twenty year sentence was appropriate. Our majority holds that the judge can vary his determination of appropriateness solely upon whether or not the defendant elects to stand trial. That makes the judge an advocate of guilty pleas, and a most powerful advocate in the bargain. I cannot agree. My disagreement, however, would not make the practice unavailable to our colleagues in the state judicial systems. Only if the practice offends the Constitution would it be forbidden. I conclude that it does.
From that point of view, which I believe to be the right one, let us review the precedents.
II. The Trial Judge and Plea Bargaining
In no way am I challenging the constitutional propriety of plea bargaining. Rather, my problems begin with the majority’s definition of plea bargaining.
Plea bargaining is a process of negotiation in which the prosecutor, trial judge, or some other official in the criminal justice system offers the defendant certain concessions in exchange for an admission of guilt.
(At 875.) (Emphasis added.)
To my knowledge, neither the Supreme Court, see, e. g., Santobello v. New York, 404 U.S. 257, 92 S.Ct. 495, 30 L.Ed.2d 427 (1971), nor this circuit, see, e. g., Brown v. Beto, 377 F.2d 950 (5th Cir. 1967), has ever so defined plea bargaining. To the contrary, participation in the plea bargaining process has been sharply limited to the accused and the prosecutor.
In Brown v. Beto, supra, Judge Wisdom, writing for the panel, concluded that “[pjroperly safeguarded plea discussions and plea agreements between an accused and a prosecutor are consistent with the fair administration of justice.” Id., at 956 (Emphasis in original.) He further concluded that “[t]he trial judge should not participate in plea discussions.” Id., at 957, citing, American Bar Association Project on Minimum Standards for Criminal Justice, Standards Relating to Pleas of Guilty, § 3.3(a) (App. Draft, 1968). The Supreme Court has *897described plea bargaining in similar language.4
The disposition of criminal charges by agreement between the prosecutor and the accused sometimes loosely called “plea bargaining,” is an essential component of the administration of justice. Properly administered, it is to be encouraged.
Santobello v. New York, 404 U.S. 257, 260, 92 S.Ct. 495, 498, 30 L.Ed.2d 427 (1971) (Emphasis added.)
There is good reason for the exclusion of the court from plea bargain negotiation. Plea bargaining has been described as “give-and-take negotiation ... between the prosecution and defense, which arguably possess relatively equal bargaining power.” Bordenkircher v. Hayes, 434 U.S. 357, 362, 98 S.Ct. 663, 667, 54 L.Ed.2d 604 (1978), quoting, Parker v. North Carolina, 397 U.S. 790, 90 S.Ct. 1458, 25 L.Ed.2d 785 (1969) (opinion of Brennan, J.).
Equal bargaining power is essential to constitutionally valid plea bargaining. Indeed, the Supreme Court has stated that “plea bargaining flows from ‘the mutuality of advantage’ to defendants and prosecutors, each with his own reasons for wanting to avoid trial.” Bordenkircher v. Hayes, 434 U.S. at 363, 98 S.Ct. at 668, quoting, Brady v. United States, 397 U.S. 742, 752, 90 S.Ct. 1463, 1471, 25 L.Ed.2d 747 (1969). Unless this mutual advantage exists the system breaks down; bargaining becomes coercion.
Of course, Jimmy Frank went to trial so there is no allegation of a coerced plea.5 Coercing a plea, however, is only one way in which a trial judge may violate the Constitution by intruding into the plea bargaining process. Here, we are confronted with another. Specifically, we must decide under what circumstances a judge may impose a harsher penalty following conviction when he has previously “proposed” an appropriate sentence in exchange for a plea of guilty. I can come to no other conclusion but that North Carolina v. Pearce, 395 U.S. 711, 89 S.Ct. 2072, 23 L.Ed.2d 656 (1969) should control the outcome.
III. The Applicability of Pearce
At the outset, the essence of Pearce must be made clear. It was recently articulated by the Supreme Court.
The Court has emphasized that the due process violation in cases such as Pearce and Perry [Blackledge v. Perry, 417 U.S. 21, 94 S.Ct. 2098, 40 L.Ed.2d 628 (1974)] lay not in the possibility that a defendant might be deterred from the exercise of a legal right, [citations omitted] but rather in the danger that the State might be retaliating against the accused for lawfully attacking his convictions.
Bordenkircher v. Hayes, 434 U.S. 357, 363, 98 S.Ct. 663, 667-68, 54 L.Ed.2d 604 (1978).
Likewise, the due process violation in this case is not based on the possibility that Jimmy Frank was deterred from exercising his constitutional right to trial, but rather on the danger that the State might be retaliating against Jimmy Frank for lawfully choosing to exercise his right to trial.
I wholeheartedly agree with the majority “that the mere imposition of a longer sentence than [the] defendant would have received had he pleaded guilty does not automatically constitute unconstitutional punishment.” (At 883.) Indeed, in Pearce the Supreme Court expressly rejected the notion that there exists an absolute constitutional bar to the imposition of a more severe sentence upon retrial. Id., 395 U.S. at 723, 89 S.Ct. at 2079.
As the Court noted, however, the existence of retaliatory motivation is extremely difficult to prove. Id. at 725, n.20, 89 S.Ct. at 2080. Hence, the Court in Pearce required that whenever a judge imposes a *898more severe sentence upon a defendant after a new trial, the reasons for his doing so must affirmatively appear on the record so that the constitutional legitimacy of the increased sentence may be fully reviewed. Id., at 726, 89 S.Ct. at 2081. The reasons for enhancement must be based “upon objective information concerning identifiable conduct on the part of the defendant occurring after the time of the original proceeding.” Id. A brief review of Pearce and its progeny vividly illustrates that the very same constitutional considerations present in Pearce are present here.
In Colten v. Kentucky, 407 U.S. 104, 92 S.Ct. 1953, 32 L.Ed.2d 584 (1972), the Court decided the applicability of Pearce to Kentucky’s two-tiered system of criminal adjudication. Kentucky allows a misdemeanor defendant convicted in an inferior trial court to seek a trial de novo in a court of general jurisdiction. The appellant in Col-ten argued that the Constitution prevented the court of general jurisdiction, after trial de novo, from imposing a sentence in excess of that imposed in the court of original trial. The court rejected the Pearce analogy. The Court found that in contrast to Pearce, the court that imposed the increased sentence after retrial in Colten was not the one whose original judgment had prompted an appellate reversal; thus, there was little possibility that an increased sentence on trial de novo could have been motivated by personal vindictiveness on the part of the sentencing judge.
In Chaffin v. Stynchcombe, 412 U.S. 17, 93 S.Ct. 1977, 36 L.Ed.2d 714 (1973), the applicability of Pearce to Georgia’s jury sentencing system was urged. Upon retrial following the reversal of his original conviction, the defendant in Chaffin, was reconvicted and sentenced to a greater term than had been imposed by the initial jury. The Court again focused on the issue of vindictiveness in holding that Pearce was not dispositive. It noted that the second jury was completely unaware of the original sentence, and thus could hardly have sought to “punish” Chaffin for his successful appeal. Moreover, the jury, unlike a judge who had been reversed on appeal, had no stake in the prior conviction or any motivation to discourage criminal defendants from seeking appellate review.
Blackledge v. Perry, 417 U.S. 21, 94 S.Ct. 2098, 40 L.Ed.2d 628 (1974), focused on prosecutorial conduct in a two-tier system like that found in Colten. It found Pearce applicable because there was a “realistic likelihood of vindictiveness” on the part of the prosecutor. 417 U.S. at 27, 94 S.Ct. at 2102. Specifically, it held that due process prevented a prosecutor’s reindictment of a convicted misdemeanant on a felony charge after the defendant had exercised his right to appeal the misdemeanor conviction and thus to obtain a trial de novo. The Court underscored the prosecution’s “considerable stake” in discouraging the appeal. Id.
In a discussion which is helpful in resolving this case the Court noted:
There is, of course, no evidence that the prosecutor in this case acted in bad faith or maliciously in seeking a felony indictment against Perry. The rationale of our judgment in the Pearce case, however, was not grounded upon the proposition that actual retaliatory motivation must inevitably exist. Rather, we emphasized that “since the fear of such vindictiveness may unconstitutionally deter a defendants’ exercise of the right to appeal or collaterally attack his first conviction, due process also requires that a defendant be free of apprehension of such a retaliatory motivation on the part of the sentencing judge.”
Id., at 28, 94 S.Ct. at 2102, quoting, North Carolina v. Pearce, 395 U.S. 711, 725, 89 S.Ct. 2072, 2080, 23 L.Ed.2d 656 (1969).
Several lessons emerge from these cases. First, under a Pearce analysis it is not necessary to demonstrate “actual retaliatory motivation.” As previously discussed, that would be virtually impossible. Rather, the court should focus on the presence or absence of a “realistic likelihood of vindictiveness.” In assessing such a likelihood, the Court has focused on those imposing the second sentence or taking the second action.
*899Where a different judge of a different court, Colten v. Kentucky, supra, or a different jury, Chaffin v. Stynchcombe, supra, impose the new sentence the possibility of vindictiveness is remote; hence no due process violation. On the other hand, where the same judge is imposing both the first and second sentence, North Carolina v. Pearce, supra, or the same prosecutor’s office writing both the first and second indictment, Blackledge v. Perry, supra, the likelihood of vindictiveness is sufficient to invoke the protection of due process.
Chaffin v. Stynchcombe, supra, draws this distinction sharply.
Thus, the jury, unlike the judge who has been reversed, will have no personal stake in the prior conviction and no motivation to engage in self-vindication. Similarly, the jury is unlikely to be sensitive to the institutional interests that might occasion higher sentences by a judge desirous of discouraging what he regards as meritless appeals.
Id., 412 U.S. at 27, 93 S.Ct. at 1983.
In the instant case the trial judge proposed a sentence of twenty years to defendant’s counsel in exchange for a guilty plea. In so doing, he impaired his impartiality. See ABA Project on Minimum Standards for Criminal Justice, Standards Relating to Pleas of Guilty § 3.3(a) (App.Draft 1968); Note, Plea Bargaining and the Transformation of the Criminal Process, 90 Harv.L.Rev. 564, 584 (1977). In effect he became a third bargainer with a well-defined position.6 He had a “personal stake” in the process being resolved his way and therefore a “motivation to engage in self-vindication.” Finally, he had an “institutional interest” in imposing punishment through the sentencing process to discourage what he may well have viewed as a meritless trial. See Chaffin v. Stynchcombe, 412 U.S. at 27, 93 S.Ct. at 1983. In short, the very same indicia of vindictiveness that were present in Pearce are present here.
It should be made clear that I am not, as Pearce did not, presuming that the trial judge was vindictive. Rather, the “realistic likelihood” that he could have been compels the rule that he must affirmatively state his reasons for imposing a harsher sentence. IV. The Majority’s Treatment of Pearce
The majority avoids Pearce by flatly declaring that it is “completely inapplicable to post-plea bargain sentencing proceedings.”7 (At 885.) If such a case were before us I might well agree. In this case, *900however, the trial judge has entered the plea bargaining process as an adversary; hence, we do not have a simple “post-plea bargaining sentencing proceeding.”
Bordenkircher v. Hayes, supra, relied on extensively by the majority, supports our position. Referring to Pearce and Perry the Court states:
In those cases the Court was dealing with the State’s unilateral imposition of a penalty upon a defendant who had chosen to exercise a legal right to attack his original conviction-a situation “very different from the give-and-take negotiations common in plea bargaining between the prosecution and defense, which arguably possess relatively equal bargaining power.” [Citation omitted.]
434 U.S. at 362, 98 S.Ct. at 667 (Emphasis added.)
The issue here is whether or not a “unilateral imposition of a penalty upon a defendant” occurred when the judge announced his “second” sentence. This case has nothing to do with the “give-and-take negotiation common in plea bargaining between the prosecution and defense.” Rather, our task is to evaluate the constitutionality of the trial judge’s entry into that process.8 In sum, if Pearce could be distinguished it certainly would not be by reference to plea bargaining situations involving only the accused and the prosecution.
V. The Record
I agree with the panel “that the judge lacked reason to increase defendant’s sen-fence over what he proposed in the plea bargaining sessions.” 605 F.2d at 915. Stated differently, there is not sufficient “objective information concerning identifiable conduct on the part of the defendant” as required by Pearce to impose a harsher sentence.9
The majority opinion discusses a number of subjects which might be taken as finding that there existed sufficient reasons, not earlier known to the trial judge, for increasing the sentence. It is clear, however, that its affirmance of the district court finally rests upon the following:
Even if the trial provided no additional evidence of character, the mere fact that Jimmy Frank refused to acknowledge his guilt and showed no willingness to assume responsibility for his conduct may have led the judge to conclude that this defendant lacked potential for rehabilitation thus justifying the imposition of a greater sentence than that offered in exchange for a guilty plea.
(At 885.)
I submit that a judge may, in some cases, find the contrite attitude of a defendant, expressed by acknowledging guilt and by other things, worthy of consideration in fixing a sentence. But the election to stand trial rather than plead guilty may not be punished by the judge who has sought to persuade the defendant to bargain away that election.10 Only if the record contains *901articulable facts as required by Pearce can a reviewing court be certain that such punishment has not been imposed in the form of an enhanced sentence.
VI. Conclusion
I agree with the majority that this case could have a potentially devastating impact upon the plea bargaining process. If a third negotiator, the trial judge, is added to the plea bargaining process its constitutional foundation will collapse. The Supreme Court has repeatedly emphasized that plea bargaining is based on negotiation between two adversaries with equal bargaining power. The addition of a third participant, especially one entrusted with ultimately sentencing the defendant, can not be reconciled with due process.
There is no precedent approving the offer of concessions, coupled necessarily with the threat of their withdrawal, by the trial judge. Where the courts have approved plea bargaining by “the State,” they involve plea bargaining by the executive. The trial judge is not “the state” in a ease styled “State v. John Doe.”
In this case, the specific constitutional danger is that a trial judge who has “proposed” a sentence in exchange for a guilty plea will punish the defendant for refusing to accept the bargain. A judge who makes such a proposal is no longer detached and neutral; he has acquired a “personal stake” in the outcome. As in Pearce, “due process ... requires that [the] defendant be freed of apprehension of ... a retaliatory motivation on the part of the sentencing judge. “Id., 395 U.S. at 729, 89 S.Ct. at 2090. Hence, whenever a judge imposes a more severe sentence upon a defendant than he originally proposed, the reasons for doing so must be based on “objective information concerning identifiable conduct on the part of the defendant” which came to his attention after the original proposal. These reasons must be recorded at the time of sentencing. As I find the record devoid of such reasons I would reverse the District Court.
. Note three to the opinion for our majority (taken from Alschuler, Plea Bargaining and Its History, 79 Colum.L.Rev. 1, 3 n.ll (1979)) suggests other bargains, not directly related to sentence, which may be struck between the executive and the defendant. While I have serious questions that the prosecutor may properly agree to “withhold damaging information from the court” the summary is otherwise illuminating. See United States v. Avery, 589 F.2d 906, 909 (5th Cir. 1979) (Hill, J„ dissenting).
. Ellison v. Georgia Railroad Co., 87 Ga. 691, 706-707, 13 S.E. 809, 813 (1891) (Bleckley, C. J.).
. A pre-trial conference, attended by the presiding Judge, Assistant District Attorney, and defense attorney, was held in Chambers. At that conference
... the prosecution systematically presented the factual merits of this case against the defendant to the Judge, together with the pertinent criminal history of the defendant (rap sheet), and, after due consideration, the presiding Judge proposed to the defense attorney that a sentence of twenty years would be imposed in exchange for a plea of guilty.
State Record, Vol. I, at 49. (Emphasis added.)
. It is puzzling that the majority quotes this language. (At 876.)
. The facts of this case do not require us to determine to what extent a trial judge may constitutionally participate in the plea bargaining process. Here, it is enough to say that his participation has triggered the necessity for the constitutional protection announced in North Carolina v. Pearce, 395 U.S. 711, 89 S.Ct. 2072, 23 L.Ed.2d 656 (1969).
. Confronted with a similar situation, Judge Weinfeld described the judicial role as follows:
A judge’s prime responsibility is to maintain the integrity of the judicial system; to see that due process of law, equal protection of the laws and the basic safeguards of a fair trial are upheld. The judge stands as the symbol of evenhanded justice ....
United States ex rel. Elksnis v. Giliigan, 256 F.Supp. 244, 254 (S.D.N.Y.1966). Clearly, when the judge adopts an adversarial position in the plea bargaining process this role is compromised.
. This statement is said to be supported by Martin v. Blackburn, 606 F.2d 92 (5th Cir. 1979), cert. denied 446 U.S. 911, 100 S.Ct. 1841, 64 L.Ed.2d 265 (1980). While Martin does support the statement, in its context, it nowhere touches upon any issue pertinent to the case here under consideration.
In Martin the petitioner pled guilty pursuant to a plea agreement after his initial conviction was reversed on appeal and he was awaiting retrial. The plea agreement was the exclusive product of negotiations between the prosecution and the defense. There were no allegations that the trial judge impermissibly participated in the plea bargaining process. Rather, Martin contended that the 24-year sentence imposed following his guilty plea was actually more severe than his original life sentence. The court rejected this argument holding that “appellant [could] receive no greater sentence than he received following the original conviction.” 606 F.2d at 93.
We have seen that, in complete accord with Pearce, a defendant who has obtained a new trial after conviction may face harsher sentencing after retrial. Thus, it is not at all inconceivable that, while awaiting retrial, a plea bargain might be reached providing for a greater sentence. It would, if appropriate, take into account facts discovered since the first sentence. The Martin court was not faced with such a situation; the bargained sentence was not so severe as the original sentence.
I am unable to see how Martin offers us any guidance at all in deciding the result of a judge’s offering to sentence a defendant to one term of confinement if convicted on a guilty plea and then imposing a greater term upon conviction after trial.
. The majority’s failure to appreciate the significance of the trial judge’s involvement in this case has resulted in a lengthy analysis of plea bargaining. For example:
Once the bargain-whether it be reduced charges, a recommended sentence, or some other concession-is rejected, however, the defendant cannot complain that the denial of the rejected offer constitutes a punishment or is evidence of judicial vindictiveness. To accept such an argument is to ignore completely the underlying philosophy and purpose of the plea bargaining system. If a defendant can successfully demand the same leniency after standing trial that was offered to him prior to trial in exchange for a guilty plea, all the incentives to plea bargain disappear; the defendant has nothing to lose by going to trial.
I have no problem with this statement-assuming the bargaining is between the accused and the prosecution. Here, the rejected offer was proposed by the judge. Hence, we have a totally different set of issues.
. It should also be noted that if the trial judge’s reasons for imposing a harsher sentence than he initially proposed were affirmatively stated, as required by Pearce, disputes concerning the record would be minimized.
. How the quoted part of the opinion is modified by a later statement is unclear. The majority says, “The sentencing which followed the trial upon the merits saw the trial judge in possession not only of more of the detailed facts of the offense itself, but of the flavor of the event and the impact upon any victims.” *901(At 885). If this means that the judge did, indeed, receive more articulable information after his 20 year “offer,” and that the Pearce standards were met, I would dissent only by reason of a different reading of the record.