dissenting: I respectfully dissent from the opinion of the court’s majority. In their application of the due process guarantee of part I, article 15 of the State Constitution, and its prohibition against fundamental procedural unfairness, they hold that evidence of a defendant’s refusal to submit to a blood alcohol test under the implied consent law (RSA 265:84 et seq.) is inadmissible, unless the police have first warned the defendant of that potential evidentiary consequence. Assuming arguendo that the majority are correct in their view of what due process requires, I would conclude that the police gave the defendant in this DWI case an adequate warning. On the basic due process issue itself, however, I would hold that no such warning is required.
There is no need to restate the facts, except to repeat that the arresting officer gave Miranda warnings before requesting the defendant to take the test. See Miranda v. Arizona, 384 U.S. 436 (1966). One of those warnings advised the defendant that if he made any statement, it could and would be used as evidence against him. See id. at 469. Because that warning was certainly adequate to advise him that a statement of his refusal to take the test would be so used, the police in this case provided the defendant with the very advice that the majority require as a condition for admitting evidence of the refusal. Even, therefore, on the majority’s own view *224of what due process requires, they have no justification for reversing this conviction.
(Lest there be anything misleading about my own reliance on Miranda, however, I should add that I do not rule out the possibility of conflict between a Miranda warning of the right to silence and the introduction of evidence that subsequent to the warning the defendant chose to remain silent as a means of refusing to submit to a blood test under the implied consent law. See Doyle v. Ohio, 426 U.S. 610 (1976). This problem is not, of course, before us, and it has no bearing on either the majority’s or the dissenters’ views of how to resolve the issue that is before us. Suffice it to say here that the potential conflict may be addressed in light of South Dakota v. Neville, 459 U.S. 553 (1983), which held that a defendant’s implied consent law obligation to choose between taking and refusing a blood test is not “compelled” within the meaning of the fifth amendment, id. at 563-64; see State v. Cormier, 127 N.H. 253, 258-59, 499 A.2d 986, 990 (1985), which it is the object of Miranda to serve. Doyle v. Ohio, supra at 617; Miranda v. Arizona, supra at 467.)
My disagreement with the majority does not stop, however, at the adequacy of Miranda warnings to satisfy their due process warning requirement, for their more fundamental error lies in holding that due process requires such a warning at all. That error appears to have its source in one or more of three separate positions that the majority adopt: about the statutory obligations imposed on the police by the implied consent law, about constitutional standards of evidentiary admissibility and fundamental fairness, and about the inferences that can reasonably be drawn from actual police behavior in cases like this one. I believe that each position is itself mistaken.
The statutory mistake lies in the majority’s interpretation of RSA 265:87,1(c), which provides that before a blood alcohol test is given the police officer must inform an arrested DWI defendant of, inter alia,, the “consequences” of a refusal to permit the test. The majority assume without explanation that the potential evidentiary use of a refusal, see RSA 265:88-a (Supp. 1986), is one such consequence to which the statute refers, and they go on to suggest that due process would be denied if a trial court were to admit evidence of a refusal when the police “officer [had] not complied with RSA 265:87.”
Much might be said about the supposed need of a State court to resort to the State due process clause to exclude evidence of the refusal, if the government really had failed to comply with a statutory requirement for validly requesting the test. Here, *225however, it is more important to explain why I believe the majority misinterpret the statute when they assume that the “consequences” of refusal that are subject to the statutory warning requirement include use of the refusal as evidence of guilt.
Of course, even the majority recognize that the statute can not be read so broadly as to require warnings about all possible effects of refusal, of whatever sort. Despite the soundness of this recognition, it nevertheless takes the majority in the wrong direction, for it leads them to employ what Justice Thayer’s dissent demonstrates to be their untenable and inconsistently applied distinction between “direct” consequences that must be warned about, and “collateral” consequences that can be ignored. Compare the majority opinion supra with State v. Jenkins, 128 N.H. 672, 517 A.2d 1182 (1986).
Instead, however, of trying to make sense of “consequences” by applying this unworkable dichotomy between the “direct” and the “collateral,” we can follow a far simpler course for determining what “consequences” the legislature had in mind, by looking to the statute as a whole. See King v. Town of Lyme, 126 N.H. 279, 284, 490 A.2d 1369, 1372 (1985) (meaning of a statute is to be determined from its construction as a whole). One immediately sees that the requirement to give advice about the “consequences” of refusing, RSA 265:87, was enacted at the same time as the original version of RSA 265:92 (Supp. 1986), which expressly describes two such consequences: a refusal will (1) preclude the test and (2) lead to a revocation of driver’s license or privilege to operate a motor vehicle. Id., subsection I. Thus, “consequences” in § 87 may most reasonably be interpreted as referring to the prohibition of the test and the revocation of the license, as each is described in § 92. It would be consistent with the legislature’s intent, therefore, to conclude that these are the “consequences” that must be warned about, and to hold that the evidentiary use of a refusal need not be the subject of warning or advice to a defendant.
This conclusion is strengthened by contrasting two other statutory provisions of the implied consent law. RSA 265:92, 11(f) (Supp. 1986) provides that the State may effect the administrative license revocation only if the police officer who requested the test also warned the defendant that his refusal would carry that consequence; but RSA 265:88-a (Supp. 1986) imposes no warning requirement when it provides that a defendant’s refusal may be used as evidence against him at trial. The contrast confirms that the legislature neither intended evidentiary admissibility to be treated as one of the “consequences” of refusal that a defendant *226need be warned about, nor assumed that it ever would be so understood. The majority’s contrary interpretation is thus unsound, and inadequate to justify today’s result even on a theory that State due process is an appropriate basis for relief from the consequences of the government’s failure to comply with a State statute.
The second mistaken view that animates the majority is, like the first, left more to suggestion than to explanation in any detail. The majority contrast the respective situations of defendants who submit to a test and those who refuse, by saying that the former probably appreciate that they will thereby provide evidence, whereas the latter probably do not understand that a refusal could be used as evidence of guilt. The majority thus assume that there is constitutional significance simply in a defendant’s likely understanding of the evidentiary potential of what he does or says. Their assumption is one that calls for some examination.
Although the majority cite no supporting authority for their position, it has an undeniably familiar ring, which, I suggest, is the echo of Miranda. It is an expansive and distorted echo, however, for the majority effectively transform the familiar and specific requirement of Miranda v. Arizona, 384 U.S. 436 (1966), into a general rule of evidence, unlimited by any reference to the constitutional privilege that Miranda was intended to serve.
It is sufficient here to recall that Miranda requirements to warn a defendant that his statement may be used against him, and to advise him of his rights to silence and the assistance of counsel, are predicated on the fifth amendment privilege against compelled self-incrimination, Miranda v. Arizona, supra at 444; see also Doyle v. Ohio, 426 U.S. 610, 617 (1976). Miranda proceeds on the theory that a defendant who is ignorant of the fifth amendment privilege may not validly waive it, and the Supreme Court’s stated object in imposing the warning requirement is to lessen the risk that a defendant will unwittingly compromise his opportunity to claim the benefit of the constitutional privilege. Id. at 467.
Enforcement of the implied consent law does not, in and of itself, raise any such risk, however. South Dakota v. Neville supra held that a DWI defendant’s obligation to choose between permitting or refusing a blood alcohol test is not a compelled response within the meaning of the fifth amendment, and our own case of State v. Cormier supra reached the same result with respect to the privilege as it is guaranteed in part I, article 15 of the State Constitution. Therefore, a police request, without more, to choose between permitting or refusing a test under the implied consent law requires no warning of potential evidentiary use under Miranda, *227and would require no such warning under the State Constitution if Miranda were a State constitutional case.
Once we understand that Miranda has no precedential relevance here, we can appreciate how significant it is that the majority do not even claim to have constitutional authority for their general assumption that evidence obtained from a defendant may not be used against him unless he understood, or at least was warned, that such use would occur. This want of authority should, indeed, come as no surprise, for we all know that courts daily admit evidence volunteered to police informants who customarily misrepresent their loyalties in order to get damaging information from defendants, who have no idea they are dealing with agents of the government, and no inkling they are providing evidence that the government will use to convict them. See State v. Lewis, 129 N.H. 787, 798, 533 A.2d 358, 365 (1987); State v. Kilgus, 128 N.H. 577, 592-93, 519 A.2d 231, 241 (1986); Hoffa v. United States, 385 U.S. 293 (1966), reh’g denied, 386 U.S. 940, 951 (1967).
With this realization that the majority can proffer no recognized and generalized constitutional basis extending beyond Miranda for requiring warnings of the evidentiary potential of a defendant’s speech, we are in a position to recognize that the invocation of due process opens up no obvious route to the relief the defendant seeks. And we are, likewise, forced to come to grips with the basic questions of just what it is that fundamental unfairness entails as a criterion for finding violations of procedural due process, and what it is that is supposed to be thus fundamentally unfair when an unwarned defendant’s uncoerced statement is used to convict him.
To understand why I believe the majority have misapprehended what a demonstration of fundamental unfairness should require, it is worth starting with a look at the small number of New Hampshire cases that have dealt expressly with the concept. In State v. Martin, 125 N.H. 672, 676, 484 A.2d 1176, 1179 (1984), this court held there was no denial of fundamental fairness in trying a defendant for DWI, merely because the State was unable to produce evidence of blood alcohol testing. And in State v. Lewis, 129 N.H. at 797, 533 A.2d at 364, we held there was no fundamental unfairness in honoring a defendant’s waiver of counsel, despite his invocation of the right to counsel during a police interview conducted two months previously.
On the other hand, we did find fundamental unfairness in Appeal of Public Service Co. of N.H., 122 N.H. 1062, 1073, 454 A.2d 435, 441 (1982), when an administrative tribunal not only adjudicated *228an issue that fell outside the stated purposes of a pending proceeding, but did so on the basis of ex parte communications to a member of the tribunal. And in Appeal of Plantier, 126 N.H. 500, 509, 494 A.2d 270, 275 (1985) we held it fundamentally unfair to entertain a proceeding for revocation of a professional license when the complainant’s nine-year delay in filing charges had prejudiced the respondent’s capacity to defend himself.
Although this body of cases is a small one, it permits at least the tentative generalizations that an adjudicatory procedure is fundamentally unfair when the procedure itself provides one party with a significant advantage, and places his opponent in a corresponding position of prejudice in the search for truth or the assessment of culpability; or when the procedure allows one party to reap the benefit of his own behavior in placing his opponent at an unmerited and misleading disadvantage. Public Service gave examples of the former when the court condemned the surprise advantage of ex parte evidence in litigating a matter never timely placed in issue; Plantier exemplified the latter, in barring a complainant from proceeding after he had lulled his opponent into forgoing the preservation of evidence that might have been employed in defense. The essence of such fundamental unfairness lies not in the mere weakness of one party’s position, but in the use of a procedure that produces the party’s disadvantage, or that allows an opponent to benefit from his own improper conduct in producing the disadvantage. While I do not mean to suggest that these examples and generalizations represent the limits of our concern to avoid fundamental procedural unfairness, I think they are enough to place the burden on the majority of this court to explain exactly what is fundamentally unfair in the procedure employed in this case, as distinguished from what the defendant finds unfortunate in the probative force of the evidence. And while, again, these examples and generalizations may not be exhaustive, they do stand as good reasons to approach the claim of fundamental unfairness now before us in the same way that the Supreme Court of the United States has done in similar cases, by asking whether the trial court admitted inculpatory evidence that a defendant was induced to provide as a result of police deception, see Doyle v. Ohio, 426 U.S. at 621 (Stevens, J., dissenting), or police trickery, South Dakota v. Neville, 459 U.S. at 565-66.
The search for police deception or trickery, or other improper conduct in this case, however, does no more than reveal the poverty of the defendant’s State due process claim. The defendant addresses the character of the police behavior with a single assertion, that *229in warning him that his license would be revoked if he refused a test, the police officer implicitly represented that a refusal would have no other consequences. This assertion, it should be recognized, states a conclusion, not a fact. It simply assumes that the police officer advised the defendant about the potential administrative revocation of his license in such a way as to suggest that he was thereby describing all possible consequences of a refusal to take the test. If this were true, of course, the defendant would have a colorable due process claim. If, for instance, the officer had made some general statement that he was about to advise the defendant comprehensively about all the legal effects of refusal, a failure to address the law of evidence might indeed suggest that a refusal would be given no evidentiary weight. Or if the officer had explicitly warned that the results of any test would be used as evidence, the failure to say more might have suggested that there was no other evidentiary consequence to consider.
The officer did not, however, indicate that he was providing a survey of all consequences, and he said nothing even touching on the law of evidence. On these facts it is simply far-fetched to argue that a defendant would reasonably have inferred that his refusal could have no possible significance beyond the administrative proceeding for revocation of his license, and it is instructive, certainly, that this defendant does not even pretend that he engaged in such a reasoning process. Prior to the first, oral implied consent warnings the defendant stated “Yes, I’m drunk, I’m cocked, I’ve had too much to drink. I shouldn’t be driving,” following which he was “boisterous and belligerent” as he rode in handcuffs to the police station. It would strain reality to suggest that the defendant may have reasoned from the premise known as “expressio unius est exclusio alterius,” and it is only slightly less of a strain to conclude generally that warning of the administrative consequence of refusing a test implicitly warrants that the refusal can have no other significance.
The implausibility of the defendant’s assertion serves to remind us again that is it not “fundamentally unfair” simply that a defendant without any constitutional privilege to refuse may have predicated his refusal on a mistake of law, or may have exercised poor judgement in refusing to take the test. For a defendant’s mistake does not necessarily imply police trickery, any more than his imprudence implies police deception. What is important is that there is not even a hint of blameworthy police conduct in this case, and in the absence of improper police activity in obtaining the *230evidence, there was no fundamental unfairness when the court admitted that evidence at trial.
The judgment of conviction should be affirmed.