dissenting, in which FAY, VANCE, COX, Circuit Judges, and HENDERSON, Senior Circuit Judge, join:
The court today holds that a new indulgence is to be tendered to the pro se litigant. We have, heretofore, construed his pleadings liberally. Today we hold that the consequences of pro se litigation shall not fall upon the pro se litigant as they do upon the counseled litigant.
This holding is made in a second round of federal habeas corpus litigation in the fed*966eral courts. It is not needed. It is not wise. I respectfully dissent.
The judicial systems of the nation and of the states have long been frustrated by unending litigation preventing finality in the administration of criminal justice. A state criminal defendant is afforded many rights to prevent an unjust conviction. Representation by counsel, trial by jury, confrontation and cross-examination of witnesses, compulsory process for the production of testimony and other evidence, presumption of innocence and the burden upon the prosecutor to prove guilt beyond a reasonable doubt are among a defendant’s protections. If convicted, a defendant may appeal; should the appeal be unsuccessful, a defendant has the right to petition the United States Supreme Court for a writ of certiorari directed to the highest court of the state. Great care is taken to see to it that a defendant is protected and that convictions are proper.
Nevertheless, collateral attack upon such convictions follows collateral attack; a judgment of guilt in a criminal case may, apparently, never be said to be final as long as the convicted defendant lives. The House Report accompanying the 1966 revisions of the procedures applicable to review by lower federal courts of habeas corpus petitions by state prisoners recognized that “[wjhile in only a small number of these applications have the petitioners been successful, they nevertheless have not only imposed an unnecessary burden on the work of the Federal courts but have also greatly interfered with the procedures and processes of the State courts by delaying, in many cases, the proper enforcement of their judgments.” H.R.Rep. No. 1892, 89th Cong., 2d Sess. 5 (1966). Courts, judges, and scholars likewise have referred to the frustrations inherent in this state of affairs. See, e.g., Kuhlmann v. Wilson, 477 U.S. 436, 450-54, 106 S.Ct. 2616, 2625-27, 91 L.Ed.2d 364, 378-81 (1986) (plurality opinion); Barefoot v. Estelle, 463 U.S. 880, 887, 103 S.Ct. 3383, 3391-92, 77 L.Ed.2d 1090, 1100 (1983); Thigpen v. Smith, 792 F.2d 1507, 1513 (11th Cir.1986); Friendly, Is Innocence Irrelevant? Collateral Attack on Criminal Judgments, 38 U.Chi.L.Rev. 142 (1970); see generally, Teague v. Lane, — U.S. -, 109 S.Ct. 1060, 1075, 103 L.Ed.2d 334 (1989) (“the principle of finality ... is essential to the operation of our criminal justice system[;][w]ithout finality, the criminal law is deprived of much of its deterrent effect.”) Federal judges appear to seek laws enacted by the legislative branch which may permit us, in appropriate cases, to reach finality. However, when such laws are given to us, we tend to reject them out of hand. Our yearning for finality does not often equal our delight in insinuating our own ideas of right and wrong into the judgments of state courts. We tip our hat to finality, but create most ingenious contrivances to avoid it. I have taken note of this tendency on some past occasions in which I have felt the need to remark that “[i]t may be that what we do speaks so loudly that no one will hear what we say.” See Rhodes v. Interstate Battery System of America, 722 F.2d 1517 (11th Cir.1984) (Hill J., dissenting); Wilson v. First Houston Investment Corp., 566 F.2d 1235, 1244-45 n. 1 (5th Cir.1978) (Hill J., dissenting), vacated, 444 U.S. 959, 100 S.Ct. 442, 62 L.Ed.2d 371 (1979).
We have, here, an example. Congress recognizes that there can be no strict application of the doctrine of res judicata to habeas corpus cases. There must not be any absolute rule of law that provides absolute finality in cases of deprivation of life or liberty through the operation of the criminal laws. That is simply because, should an innocent person be in custody or in jeopardy of the execution of a death sentence, there must be an opportunity for that person to repair to a court of justice for relief. Were that not the case, there would be no principled reason for not applying res judicata to these collateral attack cases as is done in other litigation.
Therefore, when Congress sought to give us an opportunity to reach finality in the administration of criminal justice, it did not enact the doctrine of res judicata into habe-as corpus rules, but it did provide for the application of the considerations which inform the doctrine of res judicata in collateral attack cases insofar as that can be *967done without risking the loss of the “safety valve” for the innocent, described above. Rule 9(b) of the Rules Governing Section 2254 Cases in the United States District Courts provides for the dismissal of a second or successive petition if it raises only claims that were previously brought and decided on the merits, or it raises claims for the first time that previously were available and the judge finds that the failure of the petitioner to assert those grounds in a prior petition constitutes an abuse of the writ.1 Rule 9(b), Successive Petitions, Rules Governing Section 2254 Cases, 28 U.S.C. fol. § 2254 (1982).
One need not speculate as to why these provisions were added. The reasons were clearly stated in the legislative history of the amendments to sections 2244 and 2254. Both the Senate Report and the House Report state that the purpose of the new text was to provide for “a qualified application of the doctrine of res judicata.” S.Rep. No. 1797, 89th Cong., 2d Sess., reprinted in 1966 U.S.Code Cong. & Admin. News 3663, 3664; H.R.Rep. No. 1892, 89th Cong., 2d Sess. 5-6 (1966). The Senate Report adds that the bill “seeks to alleviate the unnecessary burden [of an increasing number of petitions filed in federal court] by introducing a greater degree of finality of judgments in habeas corpus proceedings.” S.Rep. No. 1797, 89th Cong., 2d Sess. reprinted in 1966 U.S.Code Cong. & Admin. News 3663, 3664. Congress heard and responded to our expressed alarm at endless collateral litigation.
There is no need to tarry long to demonstrate that, but for an exception discovered by the court today, Mr. Gunn’s second petition would be an abuse of the writ. The only issue upon which the court orders relief is the contention that a jury instruction on intent had created a presumption of intent to kill and thus violated the rule of law laid down by the Supreme Court in Sandstrom v. Montana, 442 U.S. 510, 99 S.Ct. 2450, 61 L.Ed.2d 39 (1979). Sand-strom was decided on June 18, 1979. Two days later, petitioner was convicted in the state court. In November, 1979, petitioner took his direct appeal to the Georgia Supreme Court, but he did not claim that the jury instruction on the issue of intent had impermissibly shifted part of the government’s burden of proof. More than three years after Sandstrom was decided, Mr. Gunn brought a petition for habeas corpus in the federal district court. He again raised the issues that had been presented to the Supreme Court of Georgia in the direct appeal but did not challenge the constitutionality of the jury instructions. The district court denied relief on that petition in April, 1983. In October, 1985, more than six years after Sandstrom had been decided, petitioner in his initial state habeas corpus petition first challenged the burden shifting instruction. On April 1, 1986, petitioner filed his second federal petition for habeas corpus, raising, as his sole ground, the Sandstrom issue. The state immediately asserted that this constituted an abuse of the writ. It is this second federal petition, granted by the district court, that we consider in his appeal.
The court readily concludes that the second habeas petition raising the Sandstrom issue for the first time would constitute an abuse of the writ absent the exception it has discovered for petitioner. It states, “... if Gunn had been counseled when he prosecuted his prior habeas petition, his failure to raise the Franklin/Sandstrom claim would constitute an abuse of the writ, unless he were able to present some other “justifiable reason” for omitting the claim. E.g., Tucker v. Kemp, 819 F.2d 978 (11th Cir.1987), cert. denied, 481 U.S. 1073, 107 S.Ct. 2472, 96 L.Ed.2d 364 (1987).”
The court has discovered that a petitioner’s pro se or counseled status is crucial.2 *968In so doing, the court comes to the rather astonishing conclusion that a litigant is more responsible for what his agent fails to do than for what he fails to do himself. If a petitioner has a lawyer who acts for him in litigation, the petitioner is responsible for the negligence of the lawyer in omitting a claim in his first petition, thereby rendering his second petition abusive. However, if the litigant acts for himself and negligently omits a claim in the first petition, he in many cases would not be held responsible for that omission and may freely litigate the matter in a second petition.
The court’s opinion speaks of the equities inherent in habeas corpus litigation and, facially, states reasons that appear sufficient for giving special consideration to the pro se litigant in this context.3 If the first habeas petition in this case were ambiguous as to whether or not it was raising a particular issue, we might construe it liberally, finding that the issue was raised. However, what the court does today is hold that the consequences of litigation are not the same for one who litigates pro se and one who litigates through counsel.
Facially, it may appear “fair” to allow pro se litigants in habeas corpus, who are presumably ignorant of the law and the court procedures, to do what would otherwise be an abuse of the writ if represented by counsel. However, a brief reflection demonstrates that this will not do. The court adopts a purportedly “objective” standard through which a district court does not look to the actual knowledge of a petitioner at the time of his prior petition but rather “inquire[s] into the reasonableness of the petitioner’s prior lack of knowledge.” In inquiring into the reasonableness of a petitioner’s prior lack of knowledge, would not a court consider, in addition to pro se or counseled status, such factors as intelligence, litigation experience, and the existence and quality of assistance in determining the culpability vel non of a petitioner? Would not a petitioner who has become schooled in habeas corpus law be treated less gently than one who has no knowledge at all of collateral attack litigation? Similarly, if a pro se litigant has been acting with the guidance and assistance of a “writ writer” in the penal institution, might he not be held accountable as would one who has been represented by counsel? At any rate, would it not be necessary for the court to determine the skill of the writ writer representing or counseling the particular petitioner, comparing the writ writers’ skill in collateral attack litigation with the skills of the average attorney? If such skill measured up, would the petitioner not be held as accountable as a petitioner represented by an average attorney?
The permutations will be almost endless once we depart from treating a litigant as a litigant and begin treating a pro se and a represented petitioner differently for the purpose of ascertaining the consequences of litigation. There are so many facets of the posture of the litigant that ought to be inspected that the notion of abuse of the writ will probably disappear — just when the administration of criminal justice in this country most desperately needs it.4
I respect the right of a litigant to represent himself or herself. Pro se pleadings should be liberally construed to insure that the litigant has access. When the litigant has availed himself of this right, the conse*969quences should be the same for him and for one who is represented.
I respectfully DISSENT.5
. Rule 9(b) restates the provisions of 28 U.S.C. § 2244(b).
. In holding petitioner's neglect to be excusable, the court states that "[i]t would be odd indeed to rule that it was an abuse of the court’s discretion to find that a pro se petitioner was not reasonable in reaching the same conclusion as that reached by justices of the Georgia Supreme Court and of the United States Supreme Court." This reasoning is flawed for it leads logically to a conclusion that lawyers, and thus counseled petitioners, who failed to recognize the constitutional claim at issue in the interim between Sandstrom and Franklin also should be excused *968unless the court has concluded that lawyers should know more than justices of the Supreme Court of Georgia and the United States Supreme Court. Nevertheless, as depicted above, the court states that constructive knowledge of a Sandstrom/Franklin claim would be attributed to a lawyer, and thus to a represented petitioner. I submit that there is no real defensible basis for differentiating between counseled and pro se petitioners.
. This case, unlike the ones cited below, does not involve a court liberally construing a pro se pleading to help a pro se petitioner have the same access to the courts that he would have had if he had been represented. Golden v. Newsome, 755 F.2d 1478, 1480 (11th Cir.1985); Phillips v. Mashburn, 746 F.2d 782, 784 (11th Cir.1984); Williams v. Griswald, 743 F.2d 1533, 1542-43 (11th Cir.1984); Roberts v. Wainwright, 666 F.2d 517, 519 (11th Cir. 1982), cert. denied, 459 U.S. 878, 103 S.Ct. 174, 74 L.Ed.2d 143 (1982).
. "We have met the enemy and it is us!” (apologies to Pogo).
. I do not overlook that the court’s opinion suggests that Mr. Gunn may not have been guilty of the crime — murder—for which he was convicted. This suggestion, though, comes in that part of the opinion reaching the conclusion that the jury instruction condemned in Sand-strom/Franklin was prejudicial in the guilt-innocence trial. If the court were suggesting that this petition presents the "colorable showing of factual innocence” referred to in Kuhlmann v. Wilson, 477 U.S. 436, 454, 106 S.Ct. 2616, 2627, 91 L.Ed.2d 364, 381 (1986) (plurality opinion), I assume the opinion would have said so. No claim that an abusive petition ought to be heard because it asserts such innocence was claimed in pleadings, briefs, or argument; therefore, I take it that it need not be evaluated.