The defendant, Anthony Mavrick, appeals from his conviction on one count of possessing goods stolen from an interstate shipment, knowing the goods to have been stolen, in violation of 18 U.S.C. § 659. We affirm.

The defendant and Chester Garelli were policemen for the Chicago and Western Indiana Railroad (C & WI). On the night of March 13-14, 1978, they were assigned to normal routine patrol out of the 59th Street tower in Chicago, Illinois. Routine patrol entails covering the railroad properties, preventing thefts, and recovering merchandise found on railroad property and returning it to the proper carrier, as well as any other duties incident to police work. A patrol assigned to the 59th Street tower would normally cover an area approximately from 21st Street to 74th Street.

The government's evidence of the events of that night showed that the defendant and Garelli violated an express order that at least one of them stay at the 59th Street tower to handle the mail. Instead they went to a parking ramp at 26th Street being used by the Missouri Pacific Railroad (MoPac) and there broke into MoPac tractor trailers. Later, they picked up the merchandise taken from the trailers and loaded it into their C & WI patrol car. Most of these activities were observed by a MoPac policeman who had been assigned to watch the parking ramp because of recent thefts there. Although the MoPac policeman and another MoPac officer attempted to follow the C & WI vehicle, they were unable to do so. The policemen then contacted the FBI.

The MoPac police later found Mavrick and Garelli and followed their C & WI patrol car back to the 59th Street tower. There they saw the two in the tower parking lot transferring packages from the trunk of the patrol car to the trunks of their private automobiles. When the MoPac police drove up to make the arrest, either Mavrick or Garelli slammed the trunk of the patrol car shut. Packages recovered from the cars were determined to be those removed from the MoPac parking ramp and a search of the C & WI police car revealed a pair of bolt cutters which are not standard equipment for C & WI police and were not the property of the C & WI.

Mavrick's version of the events of that night is that Garelli and he were merely on routine patrol. They recovered property which they had discovered on the ground in the MoPac parking ramp. Since no police were at a nearby MoPac office, they returned with the merchandise to the 59th Street tower. He explained his failure to observe what the government's evidence indicated was the proper procedure for securing and reporting recovered merchandise, stating that he considered the tower unsafe because so many people had access to it and that he failed to report the theft or to turn the goods over to MoPac police because he had information that one of them was a thief. According to the defendant, he and his fellow officer were storing the packages in their private cars because they thought that was the most secure place to keep them until they could report them to reliable authorities. Thus, the gist of the defendant's defense was that he was merely securing the property pursuant to his normal duties as a policeman and that his possession of the stolen items was authorized, innocent, and in all respects lawful.

Mavrick and Garelli were charged with violating 18 U.S.C. § 659 in an indictment handed down on June 1, 1978. They retained the same attorney and pleaded not guilty. On July 20, the district court held a hearing to determine whether certain post-arrest statements which Garelli had made to law enforcement officers should be suppressed. After ruling that the statements were admissible, the court granted the defendants' motion to sever. Garelli withdrew his previous plea and pleaded guilty on August 4. The defendant's trial commenced on August 11 and concluded three days later in a jury verdict of guilty.

The defendant's brief commences with an attack on the sufficiency of the indictment and the jury instructions which, he maintains, utterly deprived him of any opportunity to present his theory of defense to the jury. He maintains that he was merely securing the goods pursuant to his normal police duties and therefore lacked the intent to convert the goods to his own use a mental state which he argues is an essential element of the crime of possession of stolen goods. An analysis of his argument requires a rather detailed discussion of the crime as defined by the statute.

The statute under which Mavrick was charged provides in pertinent part:

Whoever embezzles, steals, or unlawfully takes, carries away, or conceals, or by fraud or deception obtains from any pipeline system, railroad car, wagon, motortruck, or other vehicle, or from any tank or storage facility, station, station house, platform or depot or from any steamboat, vessel, or wharf, or from any aircraft, air terminal, airport, aircraft terminal or air navigation facility with intent to convert to his own use any goods or chattels moving as or which are a part of or which constitute an interstate or foreign shipment of freight, express, or other property; or

Whoever buys or receives or has in his possession any such goods or chattels, knowing the same to have been embezzled or stolen . . .

Shall in each case be fined not more than $5,000 or imprisoned not more than ten years, or both; but if the amount or value of such money, baggage, goods or chattels does not exceed $100, he shall be fined not more than $1,000 or imprisoned not more than one year, or both.

18 U.S.C. § 659. An examination of the statute reveals that the first two paragraphs of the statute define two distinct offenses. The first is essentially a theft offense punishing the unlawful taking of another's property. The second is a receiving and possession offense, a crime traditionally regarded as separate from theft. See United States v. Fusco, 398 F.2d 32, 36 (7th Cir. 1968). The defendant, however, relying on authorities construing the first paragraph1 maintains that an essential element of the second is the intent to convert to one's own use.2

The first paragraph defining theft explicitly provides as an element of the crime the defendant's "intent to convert to his own use." This language is absent from the second paragraph, and therefore, according to principles of statutory construction, we would ordinarily regard the omission of this element to be deliberate on the part of Congress and not the result of mistake or inadvertence. Congress knew how to require as an essential element the mental state that the defendant would have us read into the statute and it failed to do so. Compare 18 U.S.C. § 659 P. 2 With 18 U.S.C. § 641 P. 2 (punishing "(w) hoever receives, conceals, or retains (government property) With intent to convert it for his use or gain, knowing it to have been embezzled, stolen, purloined or converted") (emphasis added). Thus, the prior decisions of this and other courts have rejected the argument that the intent to convert is an essential element of the possession crime charged here. See Applebaum v. United States, 274 F. 43 (7th Cir.), Cert. denied, 256 U.S. 704, 41 S. Ct. 625, 65 L. Ed. 1180 (1921); Bloch v. United States, 261 F. 321 (5th Cir. 1919), Cert. denied, 253 U.S. 484, 40 S. Ct. 481, 64 L. Ed. 1025 (1920); Nichamin v. United States, 263 F. 880 (6th Cir. 1920) (words in indictment charging intent to convert are mere surplusage). See also United States v. Zarattini, 552 F.2d 753, 760 (7th Cir.), Cert. denied, 431 U.S. 942, 97 S. Ct. 2661, 53 L. Ed. 2d 262 (1977) (crime requires only knowledge that the goods were stolen. Specific intent is not required under the statute).

Despite the seemingly clear language of the statute and the judicial opinions rejecting the plaintiff's contention, we must consider the policy of the law not to expand common law crimes and other crimes aimed at blameworthy behavior to punish innocent conduct. See Morissette v. United States, 342 U.S. 246, 72 S. Ct. 240, 96 L. Ed. 288 (1952). Although the crimes of receiving or possessing stolen property were unknown at common law, legislative efforts to punish such conduct began as early as the 1600's in England, and virtually all American jurisdictions have defined this conduct as criminal. See W. LaFave & A. Scott, Criminal Law § 93 at 682-83 (1972). Conviction of a receiving or possession crime carries much of the moral stigma that attaches to the "infamous common-law crimes." Morissette, 342 U.S. at 252, 72 S. Ct. 240. Consequently, although most statutes punishing receiving or possessing stolen property knowing it to be stolen do not on their face contain any additional mental element, many courts have implied such a requirement. Thus, one authority states the general rule as follows:

It is not enough for guilt that one receives stolen property with knowledge that it is stolen. Otherwise, the policeman who catches a thief in possession of stolen property and who takes the booty from him in order to return it to its owner would be guilty. Some sort of a bad state of mind, in addition to the guilty knowledge, is required. This is so although this requirement is not generally spelled out in the statute defining the offense of receiving stolen property.

The necessary intent, as in the related crime of larceny, is an intent to deprive the owner of his property. The receiver's purpose is generally, of course, to deprive the owner by benefiting himself. But he is equally guilty though his purpose is to deprive the owner, not by benefiting himself but rather by aiding the thief, as by hiding the stolen property for him.

W. LaFave & A. Scott, Criminal Law § 93 at 688 (1972). Accord, M. Bassiouni, Substantive Criminal Law 331 (1978) (defining the crime of receiving stolen property as "receiving, storing, concealing, buying or using stolen property with knowledge of its character and with intent to prevent the true owner from possessing it").3

The language of the possession offense defined in 18 U.S.C. § 659 and the policy of not punishing innocent conduct, however, can be reconciled by means other than construing the statute to require that the government allege and prove an additional mental element not expressly required by the words of the statute. A defendant charged with possession can be permitted to assert his innocent purpose as a defense. The possession of one holding stolen goods who knows them to be stolen would ordinarily be expected to be purposely inconsistent with the rights of the owner. Thus, it is not unreasonable to define the offense so that the government need not allege and ordinarily need not prove the defendant's intent to deprive the owner of his property. Instead, the burden may properly be placed on the defendant to produce some evidence tending to establish that his possession of the stolen goods was innocent even though he knew the goods to be stolen. See S. 1437, 95th Cong., 2d Sess. § 1733(b) (1978) (making it an affirmative defense "that the defendant bought, received, possessed, or obtained control of the property with intent to report the matter to an appropriate law enforcement officer or to the owner of the property"); Cf. W. LaFave & A. Scott, Criminal Law § 93 at 688 n. 52 (1972) ("Where an element of the crime does not appear in the statute defining the crime, it may be held that the prosecution need not allege and prove it. Rather it is up to the defendant to raise the absence of the element as a matter of defense").

The opinion of this court in Applebaum v. United States, 274 F. 43 (7th Cir.), Cert. denied, 256 U.S. 704, 41 S. Ct. 625, 65 L. Ed. 1180 (1921), is consistent with the view which we adopt here that the defendant's purpose for possessing stolen property which he knows to be stolen should be raised as an affirmative defense rather than as an element of the crime. In Applebaum, the court rejected the defendant's attack on the indictment for failure to allege his intent to convert. The court noted, however, that "(o)f course a person should not be condemned for having innocent possession of stolen property, for example, for the purpose of turning it over to the true owner or to the public authorities." Id. at 44.4 Cf. United States v. Tompkins, 487 F.2d 146, 152 (8th Cir. 1973), Cert. denied, 416 U.S. 944, 94 S. Ct. 1952, 40 L. Ed. 2d 296 (1974) (in prosecution for possession of mail knowing it to have been stolen in violation of 18 U.S.C. § 1708 the issue of defendant's authority to possess "would properly be a defense, not an element of the crime to be proved by the Government").

Judging the indictment and the jury instructions in light of the law as detailed above, we cannot say that the defendant was utterly deprived of any opportunity to present this theory of defense to the jury.

Chester Garelli, and Anthony Mavrick, did have in their possession certain goods and chattels having a value in excess of $100.00, . . . which had been embezzled, stolen and carried away from a motor truck . . . and which were moving as, were part of and did constitute interstate shipments of freight . . . the defendants then and there knowing said goods had been embezzled and stolen. . . .

The indictment did allege the essential elements of the crime: (1) possession, (2) of goods which were stolen from an interstate shipment, (3) knowing the goods to be stolen. Since the intent to convert is not an essential element of the crime, this is not a case in which the indictment must fail for want of an allegation of an essential element of the crime not stated in the statute. See United States v. Carll, 105 U.S. 611, 26 L. Ed. 1135 (1882). The indictment, of course, is not defective merely because it fails to anticipate and negative a possible defense. Evans v. United States, 153 U.S. 584, 590, 14 S. Ct. 934, 38 L. Ed. 830 (1894).

Similarly, the jury instructions informed the jury of the essential elements of the crime with which the defendant was charged.5 The defendant complains that the trial court failed to instruct the jury on his theory of defense, but his trial attorney neither tendered any such instructions nor objected to those offered by the government, See Fed.R.Crim.P. 30, and we cannot speculate that the trial court would have refused the defense's instructions had any been offered.6 Thus, our review is limited, at best, to those defects in the instructions so basic and prejudicial as to constitute "plain error." See Fed.R.Crim.P. 52(b); United States v. Esquer, 459 F.2d 431, 435 (7th Cir. 1972), Cert. denied, 414 U.S. 1006, 94 S. Ct. 366, 38 L. Ed. 2d 243 (1973). After reviewing the instructions as a whole, we cannot say that the trial court's failure to instruct the jury Sua sponte on the defendant's theory of defense constituted plain error.

The defendant does not complain that the trial court's evidentiary rulings prevented him from presenting evidence of his good faith to the jury. Compare United States v. Anost, 356 F.2d 413 (7th Cir. 1966). Instead, he maintains that the instructions deprived the jury of any opportunity to find him not guilty, even if they believed his story that he was merely securing the goods pursuant to his police duties and intended to return them to the rightful owner. We do not believe the court's instructions so straitjacketed the jury. The instructions, for example, instructed the jury on specific intent7 even though this court has held that specific intent is not a necessary element of the crime of possession. United States v. Zarattini, 552 F.2d 753, 760 (7th Cir.), Cert. denied, 431 U.S. 942, 97 S. Ct. 2661, 53 L. Ed. 2d 262 (1977). Moreover, the instructions on the element of possession stated that "(a)n act or failure to act is knowingly done if done voluntarily and intentionally, and not because of mistake or accident or other innocent reason." We believe these instructions permitted the members of the jury to consider the defendant's explanation of his conduct and are confident that had they believed his story, their verdict would have been an acquittal. See Zarattini, 552 F.2d at 760 ("viewing the instructions as a whole, the concept of wilfulness is sufficiently expressed in order to encompass the defendants' theory of good faith"). Although an instruction more explicitly directing the jury to acquit if they found that the defendant intended to return the property to MoPac would have been preferable, we cannot say that the failure to give one in this case constituted plain error.8

The defendant for the first time now complains that his Sixth Amendment right to assistance of counsel was infringed by his retained trial counsel's concurrent representation of his codefendant, Garelli. The defendant maintains that the Supreme Court's decision last year in Holloway v. Arkansas, 435 U.S. 475, 98 S. Ct. 1173, 55 L. Ed. 2d 426 (1978), requires reversal and a new trial. Holloway, however, is inapposite to the facts of this case and, applying this circuit's established rules on joint representation, we find no error warranting reversal.

Prior to trial, the district court held a hearing on two defense motions, the first to suppress certain incriminating statements that Garelli had made to FBI agents and the second to sever the two defendants for separate trials. After ruling Garelli's statements admissible, the court asked defendants' counsel:

it appears to me that you represent both Mr. Garelli and Mr. Mavrick?

MR. STERNBERG: That's correct.

THE COURT: it appears to me that there may be a conflict, and I think rather than severing the trial, perhaps there should be another attorney representing the co-defendant.

MR. STERNBERG: I have discussed that issue with the defendants, and I don't believe that is their desire.

THE COURT: There has been quite a few cases lodged in the Supreme Court of (Sic ) the conflict question and it appears that the Court has an affirmative duty to require separate counsel in circumstances such as these. I don't know the defendant's I don't know if it can be waived by the defendant.

MR. STERNBERG: I realize that. I would think the problem can be solved by the severance.

THE COURT: No question, the severance will resolve the problem.

THE COURT: I am going to grant the severance. . . .

After being granted a severance, Garelli pled guilty. At the defendant's trial Garelli's post-arrest statements and his later guilty plea were not alluded to by the parties. Nevertheless, the defendant now faults the trial court judge for failing to warn him of the possible conflict of interest created by his attorney's joint representation of both of the defendants.

The Sixth Amendment's guarantee of the assistance of counsel contemplates that the assistance will be effective and therefore unimpaired by a lawyer who is simultaneously representing conflicting interests. Glasser v. United States, 315 U.S. 60, 68-76, 62 S. Ct. 457, 86 L. Ed. 680 (1942); United States v. Gaines, 529 F.2d 1038, 1043 (7th Cir. 1976). Nevertheless, a single attorney's representation of codefendants does not inevitably violate the defendants' constitutional rights. See Holloway v. Arkansas, 435 U.S. at 482-83, 98 S. Ct. 1173; United States v. Mandell, 525 F.2d 671, 677 (7th Cir. 1975), Cert. denied, 423 U.S. 1049, 96 S. Ct. 774, 46 L. Ed. 2d 637 (1976). And even so, "a defendant may waive his right to the assistance of an attorney unhindered by a conflict of interests." Holloway, 435 U.S. at 483 n.5, 98 S. Ct. at 1178.

The federal Courts of Appeals have split on whether the trial judge has an obligation to affirmatively question codefendants represented by a single attorney to determine whether they are cognizant of the risk of conflicting interests. See Hyman, Joint Representation of Multiple Defendants in a Criminal Trial: The Court's Headache, 5 Hofstra L.Rev. 315 (1977); Comment, Conflict of Interests in Multiple Representation of Criminal Co-Defendants, 68 J.Crim.L. & C. 226, 241-47 (1977). But see note 9 Infra. This circuit has declined to adopt an automatic rule imposing such a duty on trial judges. "(T)he Sixth Amendment rights of defendants are adequately safeguarded by imposing the duty of informing defendants of the potential dangers of multiple-client representation initially on the attorneys, as officers of the court, and by admonishing the trial judges to be watchful for indicia of conflict during trial." United States v. Mandell, 525 F.2d at 677. Holloway v. Arkansas relied on by the defendant did not alter this circuit's rule. In Holloway, counsel for the codefendants made two pretrial motions for separate counsel and informed the trial court of a probable conflict of interest. The Supreme Court held that the failure of the trial judge either to appoint separate counsel or to take steps to determine whether the risk of a conflict of interests was too remote to warrant separate counsel deprived the defendants of assistance of counsel. 435 U.S. at 484, 98 S. Ct. 1173. Of critical importance in Holloway was the duty of defense counsel to ascertain if conflicts were probable and to truthfully advise the court of such a possibility. Id. at 485-86, 98 S. Ct. 1173. Thus, upon defense counsel's representation to the court that a conflict of interests could impair the defense, affirmative trial court response is required. The Court, however, expressly reserved the question of the "scope and nature of the affirmative duty of the trial judge to assure that criminal defendants are not deprived of their right to effective assistance of counsel by joint representation of conflicting interests" in the absence of a motion for separate counsel. Id. at 483, 98 S.Ct. at 1178. Therefore, we must look to this circuit's precedent in deciding the defendant's claim. Accord, Salomon v. LaVallee, 575 F.2d 1051, 1053 n.3 (2d Cir. 1978).9

In the present case, the events occurring prior to trial are almost exactly the opposite of those that triggered the trial court's duty of inquiry in Holloway. The trial judge was sensitive to the possibility of a conflict and raised the issue Sua sponte with defense counsel. Defense counsel, as an officer of the court, stated that "I have discussed that issue with the defendants, and I don't believe (separate counsel) is their desire." Although the defendant's brief suggests that this representation is "less than unequivocal," we are loath to assume that the defendant's trial counsel failed in his professional responsibility to advise him of the dangers of joint representation. See United States v. Mandell, 525 F.2d at 677 n.9. Even then the trial court took steps to remedy the danger to the codefendants' rights by severing their trials. Thus, we believe that this case is controlled by our decision in United States v. Kidding, 560 F.2d 1303 (7th Cir.), Cert. denied, 434 U.S. 872, 98 S. Ct. 217, 54 L. Ed. 2d 151 (1977), in which defense counsel made similar representations to the trial court. We noted that:

Though the court did not inquire of defendants whether they appreciated the potential dangers inherent in (defense counsel's) joint representation of them, it seems reasonable to infer from the remarks noted above that (the attorney) did inform his clients of the dangers, but the clients did not consider them serious enough to outweigh whatever advantage they saw in joint representation. . . . It is difficult to assume that defendants were not informed nor deliberate in choosing to proceed with one counsel.

Id. at 1310. Consequently, for the defendant "to succeed on this appeal in his challenge to joint representation, it becomes incumbent upon him 'to demonstrate, with a reasonable degree of specificity, that a conflict of interests actually existed at trial.' " Id. (quoting United States v. Mandell, 525 F.2d at 677).10

In an effort to establish an actual conflict of interests, the defendant makes much of the inconsistent positions of himself and his codefendant, Garelli. The defendant maintained and still maintains his innocence. Garelli made incriminating confessions and later admitted his guilt when he pleaded guilty. Nevertheless, the severance of the codefendants' trials and the complete absence of any hint of Garelli's inculpatory statements during the defendant's trial convince us that the possibility of a conflict of interests did not materialize into an actual conflict at trial. We find no evidence that any divided loyalties caused trial counsel to sacrifice the defendant's interests for the sake of Garelli. This is simply not a case where the same counsel had to make a choice between arguing the codefendants' conflicting theories of defense to the jury. Compare United States v. Carrigan, 543 F.2d 1053 (2d Cir. 1976) (codefendants in same trial). Nor is this a case where the defendant was unable to call a favorable witness or to effectively cross-examine a government witness because his attorney also represented that witness. Compare United States v. Pinc, 452 F.2d 507 (5th Cir. 1971). Moreover, the record shows that the defendant was not inhibited from testifying on his own behalf with the aid of counsel. Compare United States v. Gaines, 529 F.2d at 1045.

An attorney's joint representation of multiple defendants, however, may affect a criminal defendant adversely not only during the jury trial on the issue of guilt, but at any stage of a criminal proceeding. Holloway, 435 U.S. at 489-90, 98 S. Ct. 1173; Geer, Representation of Multiple Criminal Defendants: Conflicts of Interest and the Professional Responsibilities of the Defense Attorney, 62 Minn.L.Rev. 119, 125-35 (1978). The defendant does not complain, and we are unable to find any reasonable basis in the record for asserting, that his trial counsel's participation in plea negotiations on behalf of Garelli was inconsistent with the attorney's duty of loyalty to him. This case does not present, for example, a situation in which one codefendant agrees to provide evidence against the other in return for an advantageous plea bargain. The defendant does complain, however, that the joint representation was inconsistent with protection of his interests during sentencing. He maintains that since Garelli had confessed, trial counsel could have argued this during the sentencing hearing before the trial court. Because the defendant, however, had consistently maintained his innocence, counsel was unable to argue repentance as a factor in mitigation of his crime. The contrast between the positions of the two defendants, he argues, made it impossible for their attorney to adequately represent both.

Once again, however, we are unable to conclude that defense counsel's efforts were inhibited or impaired by the dual representation. The sentencing hearings for Garelli and Mavrick were held some two weeks apart. The transcripts of those hearings contain no indication that defense counsel attempted to play the interests of one client off against those of the other. Nor does the record reveal that defense counsel withheld from the court any factors in mitigation which would have benefited Mavrick, but prejudiced Garelli. The defendant makes no such allegation. The defense counsel argued for leniency in both cases. Although the trial court did compare the two defendants and imposed a slightly harsher sentence on the defendant than he had on Garelli,11 this was well within his discretion, and there is no reason to believe that such a comparison would not have been made had the defendants been separately represented.

As we have noted before, joint representation of codefendants is not inevitably inconsistent with the Sixth Amendment right to effective assistance of counsel. The defendant here has simply failed to establish with reasonable specificity that an actual conflict of interests hampered his defense during trial. We conclude that the joint representation did not deprive him of his Sixth Amendment right to effective assistance of counsel.

The defendant's final argument is that he was deprived of due process by the government's cross-examination of him when he took the stand. He maintains that the government questioning of him constituted impeachment by use of his silence following his arrest and following Miranda warnings in contravention of Doyle v. Ohio, 426 U.S. 610, 96 S. Ct. 2240, 49 L. Ed. 2d 91 (1976). We view the purpose of the government's questioning somewhat differently and do not believe that it constituted error given the impression created by the defendant's testimony on direct.

The defendant's arrest at the 59th Street tower was effected by the MoPac police. About two hours later, FBI agents took him and Garelli into custody. The defendant was given Miranda warnings, but declined to answer any questions without an attorney. While testifying in his own behalf on direct, the defendant, after relating his version of the events leading up to his arrest, told the jury of his arrest by the MoPac police and his attempt to explain to them what he was doing with the property:

Q Did you attempt at the time of your arrest to explain your conduct to those officers who were taking you into custody?

A Yes, we did.

Q Did they give you an opportunity at that time to speak?

A No, they told us to shut up, and they don't want to hear it.

During cross-examination, the prosecutor, over defense counsel's objection, attempted to contradict the defendant's assertion that he was given no opportunity to explain:

Q . . . You testified I believe, on direct examination that when you were arrested by Officer Webb and Officer Pichler shortly after 5:00 o'clock, 5:10, you attempted to explain what was going on, but they would not permit you to answer.

A That is true.

Q Later on in the morning before you left that area, did you have an opportunity to explain what happened?

A No.

Q To anyone under any circumstances?

A Anyone. My attorney wasn't there.

The prosecution did not pursue the matter any further and it did not argue it to the jury during closing argument.

In Doyle v. Ohio, 426 U.S. 610, 96 S. Ct. 2240, 49 L. Ed. 2d 91 (1976), the Supreme Court held that the use of the defendant's silence at the time of arrest and after receiving Miranda warnings to impeach his exculpatory story offered later at trial violated due process. The Court offered two reasons for its constitutional rule. First, all "post-arrest silence is insolubly ambiguous because of what the State is required to advise the person arrested." Id. at 617, 96 S.Ct. at 2244. Thus, the ambiguity of post-arrest silence makes it impossible to find the defendant's silence inconsistent with his exculpatory story offered at trial. Cf. United States v. Hale, 422 U.S. 171, 95 S. Ct. 2133, 45 L. Ed. 2d 99 (1975). Second, the Miranda warnings implicitly assure the defendant that the assertion of his right to be silent will not be used against him. Consequently, it would be fundamentally unfair to allow the use of the defendant's silence to impeach his explanation later. 426 U.S. at 618, 96 S. Ct. 2240.

Nevertheless, the Court in Doyle did not establish a complete ban on all inquiry into a defendant's silence in the face of Miranda warnings. It was careful to note:

It goes almost without saying that the fact of post-arrest silence could be used by the prosecution to contradict a defendant who testifies to an exculpatory version of events and claims to have told the police the same version upon arrest. In that situation the fact of earlier silence would not be used to impeach the exculpatory story, but rather to challenge the defendant's testimony as to his behavior following arrest. Cf. United States v. Fairchild, 505 F.2d 1378, 1383 (CA5 1975).

Id. at 619 n.11, 96 S. Ct. at 2245. It is also significant that the Court chose to base its holding on the flexible requirements of due process, instead of the Fifth Amendment, See id. at 620, 626-36, 96 S. Ct. 2240 (Stevens, J., dissenting) (noting that the petitioners had argued that the impeachment had violated the privilege against self-incrimination as well as due process), even though a ban on the prosecution's use of a defendant's silence after Miranda warnings seemed contemplated by the Court in its Miranda decision. See Miranda v. Arizona, 384 U.S. 436, 468 n.37, 86 S. Ct. 1602, 16 L. Ed. 2d 694 (1966). Thus, the rule prohibiting the government's use of a defendant's post-arrest silence depends upon a balancing of defense and prosecution interests and upon considerations of fairness within the context of the truth-seeking function of trials. Compare New Jersey v. Portash, --- U.S. ----, ----, 99 S. Ct. 1292, 1297, 59 L. Ed. 2d 501 (1979) (Statements made to grand jury under grant of immunity and under threat of contempt are coerced and their use for impeachment violated the Fifth Amendment. "Here, . . . we deal with the constitutional privilege against compulsory self-incrimination in its most pristine form. Balancing, therefore, is not simply unnecessary. It is impermissible.").

In the present case, unlike Doyle, the prosecutor's questioning was not clearly directed at eliciting an admission of the defendant's post-arrest silence. The prosecutor merely asked whether the defendant was given an opportunity to explain at any time after his arrest. Although such an inquiry may ordinarily be improper because it tends to elicit testimony about the defendant's silence, here the defendant himself raised the issue of his opportunity to explain during his testimony on direct. That his testimony on direct related only to his initial confrontation with MoPac police and the prosecutor's questioning related to his later contact with the FBI is inconsequential. Having raised the question of his opportunity to explain to police officers, he "opened the door to a full not just a selective development of that subject." United States v. Fairchild, 505 F.2d 1378, 1383 (5th Cir. 1975). The defendant's arrest by the MoPac police and placement in the custody of the FBI were accomplished within a relatively short time and occurred in the same area. We cannot say that the two events were so remote that the questioning about the latter shed no light on the defendant's testimony about the former.

We view this case as similar to those decisions permitting the government to use a defendant's post-arrest silence to impeach his assertion that he had cooperated with police after his arrest. See United States v. Fairchild, supra; United States v. Conlin, 551 F.2d 534, 536-37 (2d Cir.), Cert. denied, 434 U.S. 831, 98 S. Ct. 114, 54 L. Ed. 2d 91 (1977); Stone v. Estelle, 556 F.2d 1242 (5th Cir. 1977), Cert. denied, 434 U.S. 1019, 98 S. Ct. 742, 54 L. Ed. 2d 767 (1978). These cases seem to fall squarely within the exception mentioned in footnote 11 of Doyle permitting impeachment of the defendant's version of his post-arrest behavior. Here the defendant implied that had he been given the opportunity he would have explained his seemingly illegal conduct to those detaining him. Although the defendant's testimony would not permit the government to argue that his post-arrest silence was inconsistent with his claim of innocence, we hold it did permit the government to attempt to discredit his testimony by showing that he was given such an opportunity and did not take advantage of it. We find nothing fundamentally unfair to the defendant in permitting the government to correct an impression that the defendant created by his own testimony.

The defendant calls our attention to the court's warning in United States v. Edwards, 576 F.2d 1152, 1155 (5th Cir. 1978), "that the comment upon silence of the accused is a crooked knife and one likely to turn in the prosecutor's hand. The circumstances under which it will not occasion a reversal are few and discrete. We suggest that it be abandoned as a prosecutorial technique." We join in that court's recommendation, but believe this is one of those few and discrete cases not warranting reversal. The government did not argue the defendant's silence to the jury. Its questioning was not directed toward suggesting to the jury that the defendant's silence was necessarily inconsistent with his version of the events preceding his arrest. The government merely attempted to correct the impression about the events after arrest that the defendant created by his testimony on direct. "Having voluntarily taken the stand, (the defendant) was under an obligation to speak truthfully and accurately, and the prosecution here did no more than utilize the traditional truth-seeking devices of the adversary process." Harris v. New York, 401 U.S. 222, 225, 91 S. Ct. 643, 646, 28 L. Ed. 2d 1 (1971).

AFFIRMED.