dissenting in which BLOOM, J., joins.
For the reasons so well stated by Judge Moylan in his majority Opinion, this is a “hard case.” We must be careful, however, that we do not allow a hard case to make bad law. I cannot join the majority because I believe that it is stretching the law in an unwarranted manner—in a manner that will affect thousands of other cases—simply to avoid what it perceives to be an unfairness to society in this one case. I write separately not because I necessarily disagree with Judge Salmon’s dissent, but because I do not believe it necessary to reach the Constitutional issue or the issue of whether Eiland was “available” or “unavailable” as a witness.
This appeal was reargued en banc in order to give the parties an opportunity to explain their theories and explicate the facts underlying those theories to the entire Court. The reargument served to clarify, and greatly narrow, the competing positions. The State conceded that Eiland’s refusal to *567answer questions put to him at Tyler’s trial was not based on any actual inability, physical or mental, to answer the questions, but represented simply his unwillingness to answer those questions. That concession is a legitimate one. It is the only reasonable inference that can be drawn from the record and was the necessary underpinning for the court’s finding of contempt.
The State also acknowledged that Eiland’s recorded trial testimony was not evidentially inconsistent with anything he said at Tyler’s trial. That acknowledgement necessarily served to withdraw or negate any contention that the earlier testimony was admissible because of its quality as a prior inconsistent statement. These two concessions effectively destroy the major premise for Judge Murphy’s concurring opinion.
On the basic hearsay level, the State’s position now rests entirely on the view that Nance v. State, 331 Md. 549, 629 A.2d 633 (1993), should be read to permit a prior recorded statement of a witness to be admitted as substantive evidence even if that statement is not inconsistent with the witness’s trial testimony.
I do not read Nance as supporting that position; nor can I find any other applicable hearsay exception, at least under the circumstances of this case. For that reason, I would declare the recorded statement inadmissible under the State hearsay rule and not reach the confrontation issue or the question of whether Eiland was available or unavailable at Tyler’s trial.
No one even suggests that Eiland’s recorded trial testimony was not hearsay. It was, in the words of Md. Rule 5-801(e), “a statement, other than one made by the declarant while testifying at the trial or hearing, offered in evidence to prove the truth of the matter asserted.” To be admissible, therefore, it must fall within at least one of the recognized exceptions to the hearsay rule. The statement was not offered, and its admissibility cannot be justified, under any of the express exceptions formerly recognized as part of Maryland common law that are now contained in Md. Rules 5-803 or 5-804. Nor *568can it be regarded as admissible under the “catchall” exceptions stated in Rule 5—803(b)(24) or 5-804(b)(5). Assuming that those broad exceptions existed at common law, the court made none of the requisite findings sufficient to justify admission under those provisions.
As noted, the State seeks to sustain admission of the statement solely upon an expanded reading of Nance.
Nance addressed the problem of three witnesses who (1) prior to trial, had made photographic identifications of Nance as one of the persons who shot the victim, (2) also prior to trial had given written statements to the police and testimony to a grand jury naming Nance as one of the killers and describing the circumstances surrounding the shooting, but (3) at Nance’s trial, repudiated both the identifications and the recorded statements.
The Court of Appeals treated the identifications and the broader pretrial statements separately. With respect to the pretrial identifications, the Court relied on cases such as Bedford v. State, 293 Md. 172, 443 A.2d 78 (1982), and held:
“It is well settled in Maryland that a court may admit, as substantive proof, evidence of a third party testifying as to an extrajudicial identification by an eyewitness when made under circumstances precluding the suspicion of unfairness or unreliability, where the out-of-court declarant is present at trial and subject to cross-examination, [citations omitted] An extrajudicial identification is sufficient evidence of criminal agency to sustain a conviction, even though the declarant is unable to identify the accused at trial.”
331 Md. at 560-61, 629 A.2d 633.
In Bedford, the defendant was charged with the armed robbery of two elderly victims. The victims provided a description of their assailant, assisted the police in preparing a composite picture, and subsequently made a photographic identification of the defendant Bedford. At a later suppression hearing, howeVer, they were unable to make an identification, and, as a result, they were not even asked to identify Bedford at trial. The pre-trial photographic identification was *569admitted into evidence in default of such testimony and apparently formed the principal basis of Bedford’s conviction. The issue before the Court of Appeals was whether the conviction could rest on that pre-trial identification, in light of the victims’ inability to make a judicial identification. Implicit in that issue was the assumption that the pre-trial identification was admissible as substantive evidence.
Bedford, as confirmed in Nance, is instructive in two interrelated respects. The first has to do with reliability. The extrajudicial identification was ruled admissible as substantive evidence and declared sufficient to sustain the conviction because (1) it was made under circumstances “precluding the suspicion of unfairness or unreliability,” and (2) notwithstanding their later inability to make an identification, the witnesses were in court and subject to cross-examination. The second point of interest is that, in Bedford, though not in Nance, the presumed admissibility of the pre-trial identification did not depend on any recantation by the witnesses or on the pre-trial identification being inconsistent with any testimony given at trial. There was, in fact, no recantation or inconsistency in Bedford. The pre-trial assertion was admitted on much the same basis as a past recollection recorded, and, indeed, both situations are now treated together in Md. Rule 5-802.1.
In this regard, Rule 5-802.1(c) essentially codifies the Nance and Bedford holdings. It provides, in relevant part, that a statement that is one of identification of a person made after perceiving the person is not excluded by the hearsay rule if the declarant is a witness who testifies at trial and is subject to cross-examination concerning the statement. Implicit in the rule, though not expressly stated in the text, is the prerequisite that the identification be made under circumstances precluding the suspicion of unfairness or unreliability. That gloss is necessarily imposed by Rule 5-403, allowing the exclusion of otherwise admissible evidence if the probative value of that evidence is substantially outweighed by the danger of unfair prejudice.
*570The second aspect of Nance concerned the written statements made by the three witnesses to the police and, eventually, to the grand jury. Those statements, the Court noted, “were repudiated at trial.” 331 Md. at 564, 629 A.2d 633. In most instances, the repudiation was clear and direct. The issue was whether Maryland would continue to adhere to what had become a minority view that such inconsistent statements were admissible only for impeachment purposes. The entire discussion by the Court was in the context of prior inconsistent statements, as was its ultimate holding. At 569, 629 A.2d 633, the Court stated its conclusion thusly:
“We hold that the factual portion of an inconsistent out-of-court statement is sufficiently trustworthy to be offered as substantive evidence of guilt when the statement is based on the declarant’s own knowledge of the facts, is reduced to writing and signed or otherwise adopted by him, and he is subject to cross-examination at the trial where the prior statement is introduced.”
That limitation is also explicit in Rule 5-802.1(a), which effectively codified the Nance holding.
The teaching of Nance and Bedford, as currently expressed in Rule 5-802.1 is this: To the extent that the prior out-of-court statement is simply one of identification, it need not be inconsistent with any testimony given by the declarant at trial, but it must have been made under circumstances negating any suspicion of unfairness or unreliability. To the extent the earlier statement concerns matters other than identification, it must be inconsistent with the declarant’s trial testimony.
Regrettably, in this case, the State fares poorly in both aspects. To the extent that Eiland’s own trial testimony constitutes an identification of Tyler as the actual killer, that testimony was certainly not given under circumstances precluding the suspicion of unreliability. Eiland was on trial for murder; in his first trial, he had been convicted of second degree murder and use of a handgun and had been sentenced to prison for 30 years. There was never much dispute that one or the other of them fired the fatal shots, so the only *571reasonable hope that Eiland could possibly have of escaping another conviction was to place all of the blame on Tyler, which is what he succeeded in doing. The fact that his testimony was under oath hardly suffices to wash away that compelling incentive to accuse Tyler. The identification aspect of his trial testimony was therefore inadmissible because it was given under circumstances nine months pregnant with the suspicion of unreliability.
That same unreliability would doom the penumbral aspects of his testimony as well, but even if it did not, I can find no warrant whatever for extending the second holding of Nance to include non-inconsistent statements. There is nothing in Nance to suggest such an extension, and there is nothing in Rule 5-802.1 to suggest it. Indeed, were we to construe either Nance or the rule in such an extended manner, we would be creating a new, independent hearsay exception out of whole cloth, barely a year after the Court of Appeals, on the heels of a five-year effort by its Standing Committee on Rules of Practice and Procedure, comprehensively rewrote the law of evidence in this State. We would, in addition, be creating a very serious Constitutional confrontation issue, at least in criminal cases. Because the State has rightly acknowledged that there was no inconsistency here, those aspects of Eiland’s trial testimony not relating to identification of Tyler were also inadmissible.
It is for these reasons that I dissent.