Harriet Marie Jackson v. Victor G. Walker, Warden, Louisiana Correctional Institute for Women

*141 GOLDBERG, Circuit Judge: This is an appeal by the government from the district court’s ruling granting the ap-pellee’s petition for a writ of habeas corpus. I. On July 16, 1973, two men, armed and wearing disguises, entered the home of the Magee family in Concordia Parish, Louisiana. They tied up Mrs. Magee, took money and weapons from her purse, and then seized her 10-month-old daughter. The men attempted an escape in the Magees’ car, but were chased and stopped by police. The child was recovered unharmed. The appellee, Harriet Marie Jackson, was indicted for allegedly participating in this criminal enterprise. Jackson, a domestic worker for the Magees, allegedly helped plan and perpetrate the robbery and kidnapping. Evidence at her trial indicated that she knew the two men, helped plan the burglary and kidnapping in this case, allowed the men to enter and “case” the house on two prior occasions, helped them hide in a guest house on the premises on the day of the crime, and left a certain side door unlocked as planned for the men to enter the house. On August 21,1973, a grand jury indicted Jackson for aggravated kidnapping, armed robbery, and conspiracy to kidnap. The charges carried maximum sentences of life imprisonment, 99 years, and 20 years, respectively. The state first tried Jackson for aggravated kidnapping. 1 The jury convicted her of the lesser included offense of simple kidnapping. The trial judge imposed the maximum sentence of five years imprisonment and a fine of $2,000. The Louisiana Supreme Court reversed on the ground that a statement the defendant made to police was inadmissible under the 6th Amendment. State of Louisiana v. Jackson, 303 So.2d 734 (1974). Five months later, on April 9, 1975, Jackson was indicted again, this time for the ’ crimes of aggravated burglary, armed robbery and aggravated kidnapping. 2 She was tried on the aggravated burglary charge. The jury convicted her on this charge, which carried a maximum sentence of 30 years. The judge sentenced Jackson to twelve years, with credit for time served. The conviction was affirmed by the Louisiana Supreme Court, State of Louisiana v. Jackson, 332 So.2d 755 (1976). In September, 1976, Jackson filed a petition for federal habeas corpus alleging that the second indictment and trial were the result of prosecutorial vindictiveness, a violation of due process under Blackledge v. Perry, 417 U.S. 21, 94 S.Ct. 2098, 40 L.Ed.2d 628 (1974). The district court found that the second prosecution, coming after Jackson’s successful appeal of a related conviction, created the reasonable apprehension of prosecutorial vindictiveness and thus constituted a per se violation of Jackson’s constitutional right to due process. Three months later we decided Hardwick v. Doo *142 little, 558 F.2d 292 (5th Cir. 1977). Hard-wick, which interpreted Blackledge and related cases, makes it clear that in some cases the apprehension of vindictiveness is sufficient only to establish a prima facie showing of unconstitutional vindictiveness. Upon this showing, the burden shifts to the state to demonstrate that the reason for the increase in charging was other than to retaliate against the defendant for the exercise of her legal rights. If the state fails to meet this burden, the court must find actual vindictiveness 3 and a violation of the due process clause. II. A. Prosecutorial Vindictiveness after Blackledge and Hardwick. In Blackledge the Supreme Court made clear that a prosecutor’s discretion to rein-dict a defendant is limited by the due process clause. 4 In that case the defendant, convicted of an assault misdemeanor in a state court, claimed his right to a trial de novo in a higher court. The prosecutor then obtained a superseding indictment charging the defendant with a felony, assault with intent to kill, based on the same act as the earlier charge. The Supreme Court held that the second indictment violated the due process clause. It said: A person convicted of an offense is entitled to pursue his statutory right to a trial de novo, without apprehension that the State will retaliate by substituting a more serious charge for the original one, thus subjecting him to a significantly increased potential period of incarceration. Due process of law requires that such a potential for vindictiveness must not enter into North Carolina’s two-tiered appellate process. 417 U.S. at 28-29, 94 S.Ct. at 2103. Significantly, the Court stated that it saw 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 . ., however, [is] not grounded upon the proposition that actual retaliatory motivation must inevitably exist. Rather, . ‘since the fear of such vindictiveness may unconstitutionally deter a defendant’s exercise of the right to appeal or collaterally attack his first conviction, due process also requires that a defendant be freed of apprehension of such a retaliatory motivation on the part of the [prosecutor].’ 417 U.S. 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 5 . *143 In effect Blackledge sets up a per se rule for some situations. It lays down the principle that in some situations a due process violation can be established by a showing that defendants might have a reasonable apprehension of prosecutorial vindictiveness, without a showing that the prosecutor actually had a vindictive or retaliatory motive to deter appeals. There are at least two reasons for such a per se rule. First, it is difficult to prove in court the actual state of mind of a prosecutor during his exercise of discretion. And second, reindictments that look vindictive, even though they are not, may still make future defendants so apprehensive about the vindictiveness of prosecutors that they will be deterred from appealing their convictions. However, these policies should not result in an unlimited per se rule. The due process policy must be reconciled with the countervailing policy of allowing the prosecutor broad discretion to control the decision to prosecute. Hardwick, 558 F.2d 292, 301. 6 See also U. S. v. Cox, 342 F.2d 167 (5th Cir.) (en banc), cert. denied, 381 U.S. 935, 85 S.Ct. 1767, 14 L.Ed.2d 700 (1965). In reconciling these two policies, Hardwick established a framework for the necessary limitation of Blackledge’s per se rule. In Hardwick, the defendant was originally indicted for robbing a bank and assaulting policemen during the robbery. After Hardwick’s appeal, but before his retrial, a new prosecutor obtained a superseding indictment which charged Hardwick with two additional, relatively distinct criminal actions committed during the robbery episode — robbery of a bank customer, and assault on a person whom Hardwick used as a shield in the gun battle with police. In considering Hardwick’s claims that the addition of the two new charges constituted prosecutorial vindictiveness, the court in Hardwick concluded that the showing of the reasonable apprehension of vindictiveness in its case established only prima facie proof of a due process violation. The court held that the showing transferred the burden onto the state to show that the prosecutor’s motive was not actually vindictive. The court said, “the prosecutor may rebut this prima facie proof by establishing his reasons for adding the two new charges were other than to punish a pesky defendant for exercising his legal rights.” 558 F.2d at 301. The Hardwick court distinguished Black-ledge on the ground that “Blackledge involves the substitution of a more serious charge and not the making of a decision to initiate prosecution for alleged criminal activity.” The court continued: Though these latter two events [the robbery of the bank customer and the assault on the bystander used as a shield] occurred in the same overall time interval as the acts covered by the first two indictments, they were different and distinct activities and' thus were the subjects of discretionary prosecutorial decisions which up to then had not been made. These charges were not harsher variations of the same original decision to prosecute as in Blackledge. Under this court’s decision in Cox, the court is not to interfere with the “free exercise of the *144 discretionary powers of attorneys of the United States in their control of criminal prosecutions.” 342 F.2d at 171. A fortio-ri, the discretion of a State prosecuting officer in making the decision to prosecute crime may not be controlled by a court unless it be established that his motives are in fact vindictive. The apprehension of vindictiveness which controlled the decision in Black-ledge had no effect on prosecutorial discretion because there the decision to prosecute Perry for this very assault on his fellow prisoner had already been made. On the other hand, if we were to adopt apprehension of vindictiveness as opposed to vindictiveness in fact to be the standard by which we judge whether new prosecutions for different criminal activities may be initiated, we would render the prosecutor’s discretion meaningless in every case in which a defendant is initially indicted for less than all the violations his alleged spree of activity would permit. In such a situation, it is enough that a prosecutor, who decides to add charges to a prior indictment, prove that he did not in fact act vindictively. The test is to be applied to the prosecutor’s actions rather than the defendant’s reactions. Id. at 302. At first glance Hardwick and Blackledge appear inconsistent. In Blackledge the Supreme Court found that the reasonable apprehension of vindictiveness, without a showing of actual retaliatory motive, was by itself sufficient to establish a due process violation. Hardwick held that the apprehension of vindictiveness, without a determination of actual vindictiveness, established only prima facie proof of a due process violation. The appearance of vindictiveness simply shifted the burden onto the government to prove that there was no actual vindictiveness. The reason for this apparent conflict is that in Blackledge there was no strong countervailing policy of allowing the prosecutor freedom to exercise his discretion in charging other crimes. There simply were no such other crimes in Blackledge. But in Hardwick there were, and thus the court was called upon to reconcile the competing interests of defendants and the state. 7 Thus both Hardwick and Blackledge should be seen as applying the same threshold balancing test. The threshold question is whether, for a finding of unconstitutional vindictiveness, a court will require a determination of actual vindictiveness, or just a finding of the reasonable apprehension of vindictiveness. To answer this question, a court must balance the “freedom of appeal” interest of defendants against the “freedom of discretion” interest of the prosecutor. In Blackledge the defendants’ appeal interest clearly outweighed the government’s interest in unfettered prosecutorial discretion, which in that case involved only the prosecutor’s right to reopen a previously completed exercise of discretion. The premise in Blackledge is that preventing a prosecutor from raising the charge on a single offense after the defendant’s appeal entails only a minor infringement on the exercise of prosecutorial discretion, while allowing the prosecutor to raise the charge would entail a considerable threat to future defendants deciding whether to appeal their convictions. Because of this imbalance the Court did not require a determination of actual vindictiveness to decide the prosecution was unconstitutional. In Hardwick the premise was that allowing a prosecutor to charge additional crimes in a single episode after a defendant’s appeal would also entail a considerable threat to other defendants, 8 *145 but that preventing the prosecutor from charging the additional crimes would entail a very large infringement on the exercise of prosecutorial discretion. Thus in Hardwick there was a more even balance between the interests of the government and defendants. Because of this more even balance the court required a determination of actual vindictiveness to justify the infringement on prosecutorial discretion. 9 The test can be summarized as follows. In deciding whether to require a showing of actual vindictiveness or merely a showing of reasonable apprehension of vindictiveness, a court must weigh the extent to which allowing the second indictment will chill the exercise of the defendants’ appeal rights against the extent to which forbidding the second indictment will infringe on the exercise of the prosecutor’s independent discretion. In other words, the court must weigh the need to give defendants freedom to decide whether to appeal against the need to give the prosecutors freedom to decide whether to prosecute. Framing the issue in this balancing test manner may seem an unduly complicated way to reconcile Blackledge and Hardwick. But its usefulness becomes apparent in the instant case. As we explain below, the instant case involves more infringement on prosecutorial discretion than existed in Blackledge, but less than in Hardwick. But, as we show, in our case there is less risk of a chill on due process rights than existed in Blackledge or in Hardwick. By using the balancing test set out above, we can decide the threshold question of whether a due process violation is established simply by the appearance of vindictiveness, or whether a determination of actual vindictiveness is also required. To apply the balancing test we must more fully assess the relative weights of the competing interests in this case. B. Assessing the Competing Interests. 1. The prosecutorial interest. The prosecution in the instant case is a hybrid of the prosecutions in Blackledge and Hardwick. Like Hardwick, the offenses in this case of aggravated kidnapping (charged originally) and aggravated burglary (charged the second time) relate to quite different and distinct activities. And yet, as in Blackledge, Jackson’s role in both crimes entailed the same basic conduct — i. e. planning the break-in, letting the co-defendants “case” the house, hiding them in the guest house, and then leaving the door open for them. 10 *146 In effect, our case involves a decision to initiate prosecution for a different and distinct offense which was a different and distinct consequence of the same basic conduct by Jackson. Of course a prosecutor obtaining a first set of indictments has some prerogative to reserve decisions about whether to prosecute certain additional offenses growing out of the defendant’s role in a criminal episode. To a limited extent he should also be free to adopt a “different approach to prosecutorial duty” in light of new evidence or further enlightenment. Hardwick, 558 F.2d at 301. 11 But this sort of decision by the prosecutor, while deserving more deference than the pure reopening decision in Black-ledge is not as weighty as the prosecutor’s decision in Hardwick to add on charges for relatively distinct conduct in the defendant’s criminal “spree of activities.” Hardwick, 558 F.2d at 302. 2. Defendants’ appeal interest. We must weigh against this prosecutorial interest the due process interest of minimizing the apprehension of prosecutorial vindictiveness. Here the key factor is the extent to which the circumstances of the second indictment will make defendants apprehensive about the likelihood of prosecu-torial retaliation for their own decisions to appeal convictions. Paraphrasing the court in Hardwick, the sine qua non of a prosecutorial vindictiveness claim is that the second charge is in fact harsher than the first. 12 But in this case the second indictment was actually less harsh than the first charge, even though it was more harsh than the first sentence. In a technical sense, the second set of indictments was harsher than the first set. The first indictment contained three charges with maximum sentences of life imprisonment (aggravated kidnapping), 99 years (armed robbery), and 20 years (conspiracy to kidnap). The second indictment contained three charges with maximum sentences of life imprisonment (aggravated kidnapping), 99 years (armed robbery), and 30 years (aggravated burglary). Thus on one level it appears that the prosecution simply substituted a 30 year charge (aggravated burglary) for a 20 year charge (conspiracy to kidnap). But for several reasons this is a misleading account. After the jury verdict in the *147 first trial convicting the defendant for simple kidnapping instead of aggravated kidnapping, the double jeopardy clause effectively reduced the second aggravated kidnapping charge to a simple kidnapping charge, which carried a maximum sentence of only five years. See note 2 supra. Moreover, since the chief concern is the apprehension of vindictiveness, it is important to look at which charges the prosecutor chooses for actual prosecution. Indicting a defendant on a charge no doubt threatens a defendant with prosecution on that charge. Thus the indictment might constitute a retaliatory act even though the defendant is never tried on that charge. But in examining a record for Blackledge violations, our principal concern must of course be with charges on which the defendant was actually tried. Had the prosecutor here tried Jackson for a relatively minor charge at first, and then, after a successful appeal, retried her on a more severe charge, we would be more ready to suspect unconstitutional retaliation; it would not matter very much whether the aggregate of the possible penalties in the second set of indictments was greater or less than in the first. But in this case the second charge chosen for prosecution was less harsh than the first charge prosecuted. A third factor to consider in this case is the fact that the prosecutor passed up the chance to prosecute the armed robbery charge, choosing instead the lesser offense of aggravated burglary. And a fourth factor is that, after the jury convicted Jackson for aggravated burglary, the prosecutor asked the judge to set the punishment below the maximum thirty years imprisonment. Of course none of these factors by itself necessarily defeats a claim of prosecutorial vindictiveness. But taken together they make it unlikely that there are “circumstances which suggest that the increase [in charging] is retaliation for the defendant’s assertion of statutory or constitutional rights.” Hardwick, 558 F.2d 292, 301. Similarly, these factors together reduce the likelihood that other defendants will be made apprehensive about their own appeals. The key fact is that after the defendant’s successful appeal, the prosecutor actually tried the appellee on a lesser charge. In the first trial Jackson had been charged with aggravated kidnapping, which carried a maximum sentence of life imprisonment. The jury convicted her of simple kidnapping, and the judge gave her the maximum penalty of five years imprisonment and a $2000 fine. In her second trial the government prosecuted Jackson only for aggravated burglary, a charge carrying a maximum sentence of 30 years, but this time Jackson was convicted for the full offense and was given a 12 year sentence. This sequence of events, while certainly potentially discouraging to a defendant who is considering whether to appeal his conviction, does not seem to reveal vindictiveness on the part of the prosecutor. 13 After the successful appeal, the prosecutor lowered his aim from life imprisonment to 30 years, 14 and the jury and the judge, on their own, raised the penalty from 5 years to 12. 15 *148 The Blackledge “lesson” is that “the Due Process Clause is not offended by all possibilities of increased punishment upon retrial after appeal, but only by those that pose a realistic likelihood of ‘vindictiveness.’ ” 94 S.Ct. at 2102. We agree with the Louisiana Supreme Court, 332 So.2d at 759 (1976) which found that, given the circumstances of this case, “the likelihood of vindictiveness” appears remote. We think it very unlikely that defendants elsewhere will be made more apprehensive about the possibility of prosecutorial vindictiveness by the news that in Louisiana, after losing his bid to convict a person on a life imprisonment offense, a prosecutor was allowed by the court to prosecute for an offense carrying a 30 year maximum, especially where the prosecutor decided not to prosecute a related and available 99-year offense, and asked for a sentence more lenient than the 30 year maximum upon conviction. III. Under the circumstances of this case we find a very limited due process interest balanced against a moderately weighty prosecutorial independence interest. We therefore conclude that here, as in Hard-wick, a court should not interfere with the prosecutor’s exercise of discretion unless a determination of actual vindictiveness is made. We note that the district court judge found “no evidence of actual malice or bad faith on the part of the prosecutor ” 16 The Louisiana Supreme Court also found “an absence of vindictiveness,” 332 So.2d at 759, and apparently a “legitimate reason for the increased penalty” as well. Id. at 758. 17 However, these findings by the district court and by the state court were made pursuant to the mere apprehension test in Blackledge and not the actual vindictiveness approach required in Hardwick and in this case. Because Hardwick had not yet been decided, all of the courts in the history of this case did not make, or indeed know that the law required them to make, a real effort to probe the actual intent of the prosecutor, 18 or to assess fully the legitimacy of the reasons for the prosecutor’s decision to increase the charges. 19 By showing that the prosecutor increased the severity of the charges against her by a net amount of 10 years imprisonment after she had succeeded in setting aside the original conviction, Jackson has made a prima facie case for a finding of vindictiveness. “[T]he prosecutor may rebut this prima facie proof by establishing his reasons for [increasing the] charges were other than to punish a pesky defendant for exercising [her] legal rights.” Hardwick, 558 F.2d at 301. The cause is remanded to the district court to afford the prosecutor the opportunity to present countervailing evidence. 20 *149 We leave it to the trial court’s discretion as to whether a hearing will be necessary to show why the prosecutor replaced the conspiracy charge with the more serious charge of aggravated burglary. 21 We have attempted to parse the doctrinal dilemma of prosecutorial discretion infected by vindictiveness, an attempt complicated by the wisdom of the Supreme Court in Blackledge as interpreted by us in Hard-wick. Our dilemma grows out of the competition of two policies. Vindictiveness must not be allowed to eviscerate the vindication of defendant rights. But neither may courts discredit legitimate prosecutorial discretion with rules too discreet to discern, the concrete. Prosecutorial discretion is a subtle and complex mixture of cognition, analysis, purpose, and other factors. To investigate it fairly, history, atmosphere, and psychology should be plumbed in depth. With subjectivities — of the prosecutor and of defendants present and future — so central in the controversy, careful fact-finding is critical. The burden has been placed on the state but it is ultimately the court itself which must weigh the facts and balance the warring interests of the prosecutor and the prosecuted. The judgment of the court granting ha-beas corpus relief on the basis of a mere showing of the appearance of vindictiveness is vacated. The cause is remanded to the district court for further proceedings not inconsistent with this opinion. VACATED and REMANDED. Footnotes [1] At the time of the defendant’s first trial, January 14, 1974, Louisiana law did not allow charges to be consolidated for trial unless the defendant consented. See State v. Richmond, 284 So.2d 317 (La.1973); Code of Criminal Procedure, Articles 491-493, 706. [2] As the state concedes, given that the first trial resulted in a jury conviction for simple kidnapping instead of the offense originally charged, aggravated kidnapping, the double jeopardy clause of the Constitution precludes a retrial for aggravated kidnapping. Of course the state was free to retry the simple kidnapping charge, or to try any other charge involving evidence of misconduct not essentia] to the distinction between simple and aggravated kidnapping. See, e. g. North Carolina v. Pearce, 395 U.S. 711, 719-20, 89 S.Ct. 2072, 23 L.Ed.2d 656 (1969); and Blockburger v. United States, 284 U.S. 299, 304, 52 S.Ct. 180, 182, 76 L.Ed. 306 (1932). (“[Wjhere the same act or transaction constitutes a violation of two distinct statutory provisions, the test to be applied to determine whether there are two offenses or only one is whether each provision requires proof of an additional fact which the other does not.”). [3] Hardwick and other cases speak in terms of “actual malice” or “actual vindictiveness.” In one sense these terms are misleading For a prosecution to be unconstitutional, . is not necessary that the prosecutor bear any ill will toward the particular defendant in the case. The unconstitutional motive may be simply the prosecutor’s intent to discourage other criminal appeals in the future by “upping the ante” in the current appeal, even though he feels no particular malice for the current defendant. Of course, a prosecutor may also intend to punish the current defendant for appealing. The terms “malice” and “vindictiveness” more accurately describe only the latter motive, but the due process clause proscribes both motivations. [4] The Supreme Court has never held that the States are constitutionally required to establish avenues of appellate review of criminal convictions. McKane v. Durston, 153 U.S. 684, 14 S.Ct. 913, 914-915, 38 L.Ed. 867 (1894). Nonetheless, “it is now fundamental that, once established, these avenues must be kept free of unreasoned distinctions that can only impede open and equal access to the courts.” Rinaldi v. Yeager, 384 U.S. 305, 310, 86 S.Ct. 1497, 1500, 16 L.Ed.2d 577. See also Griffin v. Illinois, 351 U.S. 12, 76 S.Ct. 585, 100 L.Ed. 891; Douglas v. California, 372 U.S. 353, 83 S.Ct. 814, 9 L.Ed.2d 811; Lane v. Brown, 372 U.S. 477, 83 S.Ct. 768, 9 L.Ed.2d 892; Draper v. Washington, 372 U.S. 487, 83 S.Ct. 774, 9 L.Ed.2d 899; North Carolina v. Pearce, 395 U.S. 711, 724-725, 89 S.Ct. 2072, 2080, 23 L.Ed.2d 656; Chaffín v. Stynchcombe, 412 U.S. 17, 24 n. 11, 93 S.Ct. 1977, 1981, 36 L.Ed.2d 714. [5] In a revealing footnote, the Court said: “This would clearly be a different case if the State had shown that it was impossible to proceed on *143 the more serious charge at the outset.” The court gave as an example a case in which an assault victim died after his attacker was convicted of assault. The prosecutor was permitted to retry the defendant for homicide. 94 S.Ct. at n. 7, 2103. [6] Hardwick states: This case presents the issue of prosecutorial vindictiveness in a setting different from Blackledge. It brings into sharp conflict (1) the right of the defendant to be free of apprehension that the state might subject him to an increased potential punishment if he exercises his right to make a direct or collateral attack on his conviction with (2) the prosecutor’s broad discretion to control the decision to prosecute. 558 F.2d 292, 301. [7] See note 6 supra. [8] One might argue that adding new charges for “different criminal activities” in the same episode is inherently less intimidating to defendants than increasing the old charge for the originally charged conduct, in part because in the former situation the prosecutor “might have made [the same selection] had defendant’s appeal been unsuccessful.” State v. Jackson, 332 So.2d at 758. But the point is that a defendant apprehensive about prosecutorial vindictiveness is simply worrying about facing harsher penalties for daring to appeal, regardless of the genealogy of those penalties in the statute books. If anything, allowing a prosecu *145 tor to heap on new offenses, as opposed to simply increasing the original ones, might cause even greater apprehension among defendants because the potential for increased punishment may be much greater when new charges are added than when old charges are increased. [9] Even under Hardwick, a court does not have to see specific evidence of the intent of the prosecutor to retaliate to determine there is actual vindictiveness. It is enough that the defendant show the reasonable apprehension of vindictiveness and that the prosecutor then fails to rebut this prima facie showing of actual vindictiveness with reasonable explanations for the increased charge. In this way Hardwick also avoids the problems of proof discussed on, page 144 supra. [10] The defendant’s conduct in the two crimes was not identical, as is obvious from the fact that different evidence would be required to show Jackson helped plan, and intended a burglary as opposed to planning and intending a kidnapping. This is not simply a question of different intents, although that difference alone would distinguish the two crimes. It is also the case that counseling compatriots about how to kidnap a child in one’s care is just not the same “conduct” as giving advice on how to steal money from a purse, even though both crimes will eventually involve a few of the same physical acts by Jackson. The elements of aggravated kidnapping under Louisiana state law, LSA-R.S. 14:44, are: Aggravated Kidnapping is the doing of any of the following acts with the intent thereby to force the victim, or some other person, to give up anything of apparent present or prospective value, or to grant any advantage or immunity, in order to secure a release of the person under the offender’s actual or apparent control: (1) The forcible seizing and carrying of any person from one place to another; or *146 (2) The enticing or persuading of any person to go from one place to another; or (3) The imprisoning or forcible secreting of any person. Whoever commits the crime of aggravated kidnapping shall be punished by death; provided that if the kidnapped person is liberated unharmed before sentence is imposed then the sentence of death shall not be given but the offender shall be sentenced to life imprisonment at hard labor. The elements of aggravated burglary under Louisiana state law, LSA-R.S. 14:60, are: Aggravated burglary is the unauthorized entering of any inhabited dwelling, or of any structure, water craft, or movable where a person is present, with the intent to commit a felony or any theft therein, if the offender, (1) Is armed with a dangerous weapon; or (2) After entering arms himself with a dangerous weapon; or (3) Commits a battery upon any person while in such place, or in entering or leaving such place. Whoever commits the crime of aggravated burglary shall be imprisoned at hard labor for not less than one nor more than thirty years. While much of the difference between these crimes involves acts by Jackson’s compatriots and not actually by her, certainly her planning and participation in one crime were not the same “conduct” as her planning and participation in the other. For example, we would not allow a jury to convict her of one on the basis of findings that her conduct violated the other. See note 2 supra. [11] See note 20 infra. [12] 558 F.2d 292, 299-300. In Hardwick the observation was made in terms of vindictiveness in resentencing. This is not to say that in all cases where the prosecutor charges in the second trial a lesser offense than he charged in the first, that there is per se no prosecutorial vindictiveness and no violation of due process. Prosecutorial vindictiveness violates due process even where it creates absolutely no apprehension of vindictiveness, and does not chill any defendant’s decision to appeal. This is because vindictiveness is an improper factor in the prosecutorial process. But for a violation of due process to be found under these conditions, actual vindictiveness must be shown, not merely inferred. [13] This does not mean that the second trial could not have been a vindictive act under any circumstances. For example, there might be a case in which the original charge was so spurious or very likely excessive that the jury had little choice but to acquit on it and convict, if at all, on a lesser included offense. In such a situation we might still find vindictiveness when the prosecutor reindicts on a charge less harsh than the original charge, but more harsh than the realistic lesser included offense in the original charge. But in our case it appears that the state had more than a colorable case of aggravated kidnapping, and the original aggravated kidnapping charge was made in good faith. Under these circumstances, the fact that the offense charged at the second trial was less severe than that charged at the first does carry some, albeit not conclusive, weight. [14] Actually the prosecutor’s aim in the second trial was even lower than this. He recommended to the judge that the defendant be sentenced to less than the maximum thirty years. [15] Jackson does not argue that there was sentencing vindictiveness by the judge or jury. [16] The district court did not hold an independent evidentiary hearing on this question. It decided “on the basis of the State Court record, which provides petitioner with the equivalent of a full and fair evidentiary hearing.” [17] In making this finding, however, the court cited only a legitimate reason for the prosecutor’s decision to obtain new indictments — i. e. the need for indictments from a new grand jury composed of both men and women, since recent changes in Louisiana law cast doubt on the legitimacy of the original grand jury because it was all male. This “reason” does not explain the prosecutor’s decision to substitute in the indictment a 30-year crime for-a 20-year crime. 332 So.2d at 758. [18] The clearest example of the limited nature of the inquiry into the actual intent of the prosecutor appears in the hearing in the Louisiana trial court on Jackson’s motion to quash the indictment on Blackledge due process grounds. In that hearing Jackson’s lawyer stated: “[Sjince the requirement is not actual vindictiveness or retaliation, [the prosecutor’s] admission that he has no such motive can be taken as true.” Exhibit “B” to Record on Appeal, page 63. [19] See note 17 supra. [20] The court in Hardwick listed several ways in which an increase of charges may be constitutionally justified. It stated: An increase in the severity or number of charges if done without vindictiveness may be easily explained. For example, evidence of the additional crimes may not have been obtained until after the first indictment or information is filed, or the additional crime *149 may not be complete at the time charges are first brought. And a prosecutor may, without explanation, refile charges against a defendant whose bargained-for guilty plea to a lesser charge has been withdrawn or overturned on appeal, provided that an increase in the charges is within the limits set by the original indictment. Other explanations which would negate vindictiveness could include mistake or oversight in the initial action, a different approach to prosecutorial duty by the successor prosecutor, or public demand for prosecution on the additional crimes allegedly committed. The list is intended to be illustrative rather than exhaustive. 558 F.2d at 301. The prosecutor in the instant case has offered some explanations for his decision to increase the charges. In his brief and in oral argument before this court, the prosecutor contended that he had decided to begin the prosecution by asking the first grand jury to charge the defendant with the harsher offenses embedded in her conduct — kidnapping and armed robbery. He meant to reserve the option, and to delay his decision whether, to return to a grand jury later for charges on the lesser offenses embedded in the defendant’s complicated course of conduct and criminal involvement. This explanation is given some additional plausibility by the fact that, at the time of the first set of indictments, Louisiana state procedural law required separate indictments and separate trials of each offense charged. (See note 1, supra.) The prosecutor also makes a modest claim that he was influenced by the discovery of new evidence concerning the defendant’s leaving a door open for her compatriots. While this evidence does seem cumulative, it could conceivably influence a prosecutor to decide that a burglary prosecution was a more appropriate approach than a kidnapping charge. The trial court should consider these and any other explanations offered by the prosecutor to decide whether the prima facie case is rebutted. [21] See 28 U.S.C. § 2254(d). Considering note 18 supra, we think it likely that the criteria in § 2254(d)(2) and (3) establishing the need for a federal court hearing might be met.