concurred, making the plurality, on the basis that the exclusionary rule should not be further extended.9

It is arguable that we should adopt an earlier point, e.g., the arrest, by which to define the commencement of “all prosecutions” as that term is used in Article I, section 11, of the Oregon Constitution, but we conclude that the formal charge is the legal event which commences the obligation of the state to provide counsel. Functionally, there are no legal procedures for the appointment of counsel for indigents prior to a formal charge being brought and, as we observed in Scharf, 288 Or at 456 n 4, the decision to commence criminal rather than administrative proceedings is commonly not made until after a breathalyzer test is refused or taken and the result known. See Prideaux v. State Dept. of Public Safety, 310 Minn 405, 247 NW2d 385, 388-389 (1976). Hence, under Article I, section 11, we concur in the reasoning of Kirby that the right to counsel “in all criminal prosecutions” includes critical *805stages of the prosecution subsequent to indictment or other formal charge. Because this defendant had not been charged, no right to counsel under either constitutional provision had yet attached.

C. Liberty

Finally, if any right or interest was denied defendant by the discouragement of his request for counsel, it was that to which we referred in Scharf as “the long established and well-known right of any arrested person to call an attorney,” 288 Or at 460. To know the extent of that personal right and the limitation imposed upon the state by its existence, and in order to determine the effect of that right in individual cases, it is necessary to identify its theoretical basis. However “long established and well-known” the right of an arrested person to a phone call may be, a survey of the caselaw and authorities fails to yield any helpful discussion of the right other than in the context of helpful discussion of the right other than in the context of evidence gathering or representation during the prosecution as we discussed above. Nor have counsel been able to articulate a source of such a right. Yet any practice so traditionally and rigorously observed must have roots deep in the law. This case is unusual in that there may be an evidentiary effect to defendant’s request to call a lawyer and hence we are called upon to give some definition to the rights of communication of an arrested person for application to this case.

The usual analytical focus on a right to counsel is misleadingly narrow. The ultimate issue in this case is one of liberty, not safeguards. It is axiomatic from our organic concept of constitutioal government that every person’s liberty is complete except as the people have granted to themselves collectively (i.eto the government) the power to restrict individual liberty. The idea is as fundamental as the Social Contract and it is embodied in the Fourteenth Amendment to the United States Constitution which forbids states to “deprive any person of * * * liberty * * * without due process of law.” Thus, for example, we are free to cross streets at will, but that freedom is subject to lawful governmental restriction as to time and place to serve public safety and traffic needs because we, the people, have given police power to the government to regulate liberty in *806such ways as have a reasonable tendency to accomplish those social objectives.

Our constitutionally assured liberty is not limited to the rights itemized in the Bill of Rights such as those we discussed above. Rather, the liberty referred to by the Fourteenth Amendment, in the absence of lawful governmental restriction, is general in scope and variety. Justice Harlan expressed the idea well:

“ * * *[T]he full scope of the liberty guaranteed by the Due Process Clause cannot be found in or limited by the precise terms of the specific guarantees elsewhere provided in the Constitution. This ‘liberty’ is not a series of isolated points pricked out in terms of the taking of property; the freedom of speech, press, and religion; the right to keep and bear arms; the freedom from unreasonable searches and seizures; and so on. It is a rational continuum which, broadly speaking, includes a freedom from all substantial arbitrary impostions and purposeless restraints * * * and which also recognizes, what a reasonable and sensitive judgment must, that certain interests require particularly careful scrutiny of the state needs asserted to justify their abridgment. * * *” Poe v. Ullman, 367 US 497, 543, 81 S Ct 1752, 61 Ed 2d 989 (1961) (opinion of Harlan, J., dissenting from dismissal of appeal) (citations omitted).10

We believe that analysis to be sound and to be applicable to this case. Defendant’s freedom to call a lawyer before deciding to submit to breathalyzer testing was not safeguarded in this situation by the Sixth or Fourth Amendments, but, under the Fourteenth Amendment, his freedom to do so could not be foreclosed or deferred unless *807the police were authorized to do so. Defendant’s liberty to communicate as he chose was to be free from “purposeless restraints,” but subject to lawful restraints. Therefore, we look next to the magnitude and nature of the liberty and to whether the record discloses a lawful purpose for the denial of that liberty.

The opportunity of an arrested person to promptly communicate beyond confinement is not necessarily of the same magnitude of liberty as marital and procreational privacy which the United States Supreme Court held could only be restricted for “compelling” public interests, see Griswold and Roe, supra n 10. Neither, however, is it a minimal thing beneath the notice of the law. The everyday freedom to communicate with others takes on greater importance when one is in the enforced isolation of police custody. Communication may be the means to security release, advice, reassurance of one’s family or associates, or professional assistance. For at least these reasons, allowance of a telephone call following arrest has become traditional and incommunicado incarceration is regarded as inconsistent with American notions of ordered liberty. Freedom of an arrested person to communicate is a significant and substantial liberty which may only be officially restricted if there is legal authority to do so.

As we observed above, the authority of the police to enforce the traffic laws carries with it the authority to do all things which have a reasonable tendency to accomplish that responsibility which are not otherwise prohibited by law. Executive actions such as traffic control, detention for license checks, arrests, searches, jailing, etc., necessarily restrict some aspect of personal liberty to the degree reasonably required for the performance of that lawful act, but they are nevertheless permissible if they are within the lawful authority of the police agency to enforce the traffic laws.

The police may lawfully restrict the freedom of an arrested person to communicate to the degree reasonably required for the performance of their duties. For example, where the police are authorized to seize “highly evanescent evidence,” see Heintz, 286 Or at 248, and delay caused by an attempt to call counsel would impair their ability to *808effectively do so, they may require that the arrested person’s exercise of the freedom to call be deferred until after completion of the seizure.

Here, there was no showing that the time required to make a call would have reduced the efficacy of the breathalyzer test. Indeed, the police allowed about two hours to transpire between the arrest and the breathalyzer request, which suggests a lack of urgency. Moreover, Oregon Administrative Rule 333-13-020 (1972), which regulates the manner of breathalyzer testing, requires that the person be under observation for 15 minutes prior to the test. The record discloses no reason why a call on an available telephone during the observation period would impair the evidence gathering process, assuming the arrested person did not object to observation during the call. Nor is there evidence of the existence of what we referred to in State v. Haynes, 288 Or 59, 70, 602 P2d 272 (1979), as “the practical necessities of custody” which might have justified deferring defendant’s call to counsel. Moreover, in the absence of any such circumstances, the absolute advice on the Oregon State Police form that “[a]ny request for a delay [to have a lawyer present] will constitute a refusal,” see n. 1, supra, was not legally correct in this case. In sum this record provides no support for a holding that the police were authorized in the performance of their duties to restrict defendant’s freedom by threatening adverse consequences if defendant telephoned counsel.

We emphasize that our holding is based upon an unauthorized restriction of personal liberty, and not upon the denial or violation of a specific right enumerated in the Bill of Rights. That distinction is material. A denial of a right, e.g., the right to counsel at a critical stage of the prosecution, may occur if there is no advice, provision for the indigent, or express, knowing and voluntary waiver of that right. Freedom from interference with liberty, however, is different from entitlement to a right. The state is barred from unlawfully restricting liberty, but it is not obliged to itemize to a person all the actions he is free to take nor to provide an indigent person with the means to exercise his freedom as a precondition to the state’s taking action. Here there was no failure of the police to perform *809any obligation to protect defendant’s right to counsel; rather, under these circumstances, there was only an unauthorized restriction of defendant’s freedom to call counsel.

III. EXCLUSION

We observed in the context of self-incrimination in State v. Haynes, 288 Or at 71:

“ * * * No one so far has suggested that interference with an arrested person’s access to a lawyer, however improper and subject to other remedies, would itself lead to a reversal of a subsequent conviction if defendant in fact said nothing and no evidence was obtained as a result nor other harm done to his eventual defense. Thus it is not a generalized right to counsel that the decisions we have quoted enforce but, more concretely, the derivative right to the benefit of counsel’s efforts to forestall involuntary and incriminating disclosures. * *

Exclusion of evidence is a practical device intended to deter agents of the state from acting beyond constitutional limitations. As we recently observed in State v. Quinn, 290 Or 383, 397, 623 P2d 630 (1981):

“* * * The device of excluding trustworthy evidence from the factfinding process in order to serve higher purposes ‘is a needed, but grudgingly taken medicament; no more should be swallowed than is needed to combat the disease.’ Amsterdam, Search, Seizure, and Section 2255, 112 U Pa L R 378, 389 (1964).”

We have often applied the exlusionary rule in the context of denials of rights enumerated in the Bill of Rights, but this case is our first occasion to consider exclusion for deprival of liberty assured by the Due Process Clause.

We decline to require exclusion of the evidence for reasons arising from this record and also for reasons which transcend this particular case. This record is insufficient to establish the fact of a causal (rather than merely sequential) relationship between the request to call a lawyer and the obtaining of the breath sample. Defendant’s request for an opportunity to talk to a lawyer was general. There is no evidence that he had a lawyer or that lawyer was available, or of any other fact from which it can be inferred that if defendant had been given an opportunity, it was reasonably likely to have been fruitful. The evidence shows no *810indication that if defendant had been given the telephone and a few minutes to use it, he would have secured legal advice and, if so, whether the advice would be to take the test or not.11 A causal relationship between the official impropriety and the giving of evidence must be established by one who seeks to exclude the evidence on the ground that it is a result of that impropriety. State v. Quinn, supra, and Wong Sun v. United States, 371 US 471, 83 S Ct 407, 9 L Ed 2d 441 (1963). This record, at best, would allow us to do no more than speculate as to whether an opportunity to call a lawyer would have caused an evidentiary result.

On the broader issue, we find no case requiring or not requiring exclusion for deprival of constitutionally protected liberty interests. We begin our scratch analysis by observing that there is not just an exclusionary rule, there are several exclusionary rules, each with its own purposes and requirements. As we said in Scharf:

“Decisions on admitting or excluding improperly obtained evidence are not instances of a single ‘exclusionary rule.’ Rather, they depend on whether the premise of the impropriety was a law addressed to the manner of obtaining or using the evidence or a law protecting some unrelated interest. * * *” 288 Or at 461, n 10.

Although the rules differ, it is appropriate to review them briefly to discern principles which may guide us in determining whether exclusion of evidence obtained following a restriction of liberty must be excluded.

The exclusionary rule is applied absolutely to the fruits of violation of the Fifth Amendment. Miranda v. Arizona, 384 US 436, 86 S Ct 1602, 16 L Ed 2d 694 (1966). This is so because one’s right not to be convicted upon one’s own compelled testimony is absolute. It is not subject to moderation due to competing policies, situational variables or considerations of reasonableness. Because liberty interests are not absolute, see Section II C, above, Fifth Amendment law does not provide an apt analogy.

*811Nor is Sixth Amendment law, as it applies to evidence gathering, particularly apt. Massiah v. United States, supra, a Sixth Amendment case regulating interrogation in the absence of counsel, may now be subsumed in post-Miranda Fifth Amendment law, see e.g., Brewer v. Williams, 430 US 387, 398, 97 S Ct 1232, 51 LEd2d 424 (1977). Moreover, the lineup cases, Wade and Gilbert, supra, limit exclusion to evidence obtained solely during critical stages as an aspect of the right to confront witnesses. See also Kirby v. Illnois, supra. The reasons for exclusion in these cases do no exist here.

Fourth Amendment cases are based on principles of reasonableness and competing policy considerations. Hence they are of greater analogical significance in this case. The source of the federal exclusionary rule was Weeks v. United States, 232 US 383, 34 S Ct 341, 58 LEd 652 (1914), in which the United States Supreme Court declined to be a party to Fourth Amendment violation by allowing use of evidence obtained. By the time of Mapp v. Ohio, 367 US 643, 81 S Ct 1684, 6 LEd2d 1081 (1961), the underlying theory of exclusion had changed somewhat. The court took a pragmatic, rather than analytical, approach. It observed from the cases over the years that violations of privacy were common, that the states had found no alternative method of deterring such violations and that exclusion was therefore required for that purpose.

Since Mapp, judicial emphasis has varied somewhat, but the United State Supreme Court adheres to the view that deterrence is the dominant purpose of the Fourth Amendment exclusionary rule, but that the rule need not be applied so as to achieve maximum possible deterrence. Indeed, in United States v. Salvucci, 448 US 83, 100 S Ct 2547, 65 L Ed 2d 619 (1980), the court opted for a narrow application of the exclusionary rule because a broader application was not necessary to achieve deterrence. This theory may allow for narrowing the rule by non-exclusion where the police violate privacy rights despite their good faith intent to comply with the Fourth Amendment, United States v. Williams, 622 F2d 830, 840-848 (5th Cir 1980).

Prior to Mapp, this court never had occasion to adopt an exlusionary rule because it had never found a *812search or seizure to be unlawful. In dicta, the court stated that if the occasion ever arose, it would probably apply the federal rule for the federal reasons, State v. Laundy, 103 Or 443, 204 P 958, 206 P 290 (1922), see also State v. Hoover, 219 Or 288, 347 P2d 69 (1959), but it did not formulate a separate or alternative theory which should guide our hand today.

Subsequent to Mapp, this court has purported to apply and adopt federal search and exclusionary law. See, e.g., State v. Florance, 270 Or 169, 527 P2d 1202 (1974). It has also excluded evidence for violation of a statute which protects privacy, State v. Valdez, 277 Or 621, 561 P2d 1006 (1977), but not for violation of statutes which promote safety and accountability in search and seizure procedures, State v. Valentine/Darroch, 264 or 54, 60, 504 P2d 84 (1972), and State v. Cortman, 251 Or 566, 571, 446 P2d 681 (1968), cert den 394 US 951 (1969). These cases provide no analytical help beyond what is found in federal caselaw.

The empirical and pragmatic approach in Mapp v. Ohio is consistent with an approach expressed in State v. Shipley, 232 Or 354, 375 P2d 237 (1962). There we reconsidered whether to adopt the federal McNabb-Mallory rule12 which holds that statements obtained during a period of noncompliance with the statutory requirement that an arrested person be brought promptly before a magistrate, must be suppressed even though voluntarily given. In Shipley, we concluded that the statute requiring a hearing without delay had been violated and that a voluntary confession had been elicited during the delay. We declined to adopt a rule of suppression, however, because the statute did not require it and because the confession was voluntary. Two judges dissented. Of significance, however, and most similar to the rationale in Mapp, was the observation of Goodwin, J., concurring, that we should not automatic ally adopt an exclusionary rule. Rather, if the event were repeated and became a practice, we would reconsider adopting an exclusionary rule in order to deter future violations of law. As he put it:

*813“If those primarily charged with the duty of enforcing the law are unwilling or unable to discharge their duty in this respect, then the courts should not shrink from their duty. * * *” 232 Or at 366.

There are parallel considerations in this situation. The trustworthiness of the evidence is unaffected by the police conduct. The form used by the police implies an improper pattern of past conduct, but we may reasonably suppose that the form is the result of a good faith misunderstanding of the law rather than from an official intention to evade the law’s requirements or to hold arrested drivers incommunicado.13 We may expect that the clarification of law in this opinion will be sufficient to cause a change in police practice and deter future similar conduct without the necessity of creating a new exclusionary rule. If repeated violation occurs, then as the United States Supreme Court did in Mapp and Justice Goodwin warned in Shipley, we may consider resort to the exclusionary rule as a necessary means to deter future violations. For now, however, as we phrased it in Quinn, the disease is not so advanced that we need to swallow more of the “ ‘grudgingly taken medicament.’ ” The order of suppression is reversed, and the evidence of defendant’s breath test may be used at trial.

Reversed and remanded for trial.

specially concurring.

I concur in the result reached by the opinion by Tanzer, J. I question, however, its holding that the right of a person arrested to call an attorney is a constitutional “liberty” under the Fourteenth Amendment of the Constitution of the United States. No such contention was made by defendant in this case. I agree that a person arrested has a right to call an attorney, but am not prepared to hold at this time that such a right is a constitutional “liberty” under the Fourteenth Amendment.

With respect to the vigorous dissent by Linde, J., I agree with much of its criticism of that constitutional *814analysis in the opinion by Tanzer, J. As pointed out by that dissent, however, the previous decision by this court in State v. Scharf, 288 Or 451, 605 P2d 690 (1980), was not based upon constitutional grounds, but upon the grounds that the defendant in that case had a right to call her attorney under existing Oregon law, regardless of federal constitutional law, and that it follows from this violation of Oregon law that the results of the breathalyzer test must be excluded.

I agree with the decision in Scharf on the first ground, but disagreed, and still disagree, with the second ground of that decision. In my view, the exclusionary rule should ordinarily be applied only for violations of constitutional rights and should not ordinarily be applied for violations of statutes or on other non-constitutional grounds.1 See State v. Valentine/Darroch, 264 Or 54, 66-69, 504 P2d 84 (1972). See also State v. Bishop, 288 Or 349, 605 P2d 642 (1980); State v. Valdez, 277 Or 621, 629, 561 P2d 1006 (1977); State v. Cortman, 251 Or 566, 571, 446 P2d 681 (1968). It is also my view that the exclusionary rule is a rule which should be strictly construed in cases such as this. Driving while under the influence of intoxicants is an offense which is the cause of more deaths than any other crime or offense.

To admit into evidence the results of a breathalyzer test taken without advice of counsel is no more unreasonable, in my opinion, than to admit into evidence the results of a blood alcohol test from blood taken from an unconscious driver in a hospital, as held to be admissible in Breithaupt v. Abram, 352 US 432, 77 S Ct 408, 1 LED 2d 448 (1957). See also State v. Heintz, 286 Or 239, 594 P2d 385 (1979). For these reasons, I concur in the result reached by the majority.

dissenting.

I have joined in Justice Linde’s dissent, but I wish to elaborate on a point only lightly touched upon in his dissent.

In State v. Scharf, 288 Or 451, 605 P2d 690 (1980), we rendered a decision which was primarily a matter of interpretation of statutes of the State of Oregon. This court has specifically held that its interpretation of a statute becomes “a part of the statute as if written into it at the time of its enactment.” State of Oregon v. Elliott, 204 Or 460, 465, 277 P2d 754, cert. denied 349 US 929, 75 S Ct 772, 99 L Ed 1260 (1955).

Since the decision in Scharf, the legislature has met in regular session and failed to take any action which would indicate displeasure with that decision. Ordinarily, I would be loath to consider legislative failure to act in response to a judicial decision as having much weight in determining whether the Legislative Department of state government approved our interpretation of a statute, for mere inaction by the legislature may result from simple inattention or the pressures of other legislative business, but we do not have here a case of mere inaction.

We have, rather, the situation in which the matter has been brought to the attention of the legislature by a concerned governmental agency, and a refusal by the legislature to overturn the result of our decision in Scharf. The Department of Transportation, established as a part of the Executive Department of Oregon state government, Or Const. Art III, § 1, ORS 184.615, caused a bill to be “presession” filed in the 1981 regular legislative session. The obvious purpose of the bill, House Bill No. 2021, was to overturn the decision of this court in Scharf that the evidence was to be excluded. Rather than waste space by setting forth the bill in its entirety, I quote the summary printed on the bill:

“Permits use of chemical test results, administered pursuant to the Implied Consent Law, as evidence regardless of whether or not the driver was given opportunity to obtain advice of counsel.”

*816This bill received the immediate attention of the legislature. That body convened on January 12, 1981, and the bill was referred to the House Committee on Judiciary on January 15. On February 16, a subcommittee held a hearing on the bill. On April 15 under House rules the bill was tabled in committee for want of affirmative action. On April 21, the other subcommittee of the House Judiciary Committee held a hearing on the bill despite the fact that it had already been tabled. On April 30 during a work session the bill was taken from the table, considered, and again tabled in committee.

We see, therefore, that the legislature, acting according to the rules and methods which it has adopted for carrying out its functions and responsibilities, rejected the Executive Department’s express request to overturn our holding in Scharf that the evidence must be excluded. I would also note that the 1981 session of the legislature otherwise devoted a great deal of attention to laws relating to driving while under the influence of intoxicants and left unchanged our decision in Scharf

Given both this legislative special attention to the rule of Scharf and that body’s general concern with closely-related laws, I conclude that the legislature’s inaction here should be viewed much the same as the situation in which a legislative body, while amending or revising a particular body of law, has left the challenged section alone.

The situation in the case at bar is much like that before the United States Supreme Court in United States v. South Buffalo R. Co., 333 US 771, 68 S Ct 868, 92 L Ed 1077 (1948), where the Supreme Court was faced with a challenge to its prior decision interpreting a part of the Interstate Commerce Act in a case entitled United States v. Elgin, Joliet & Eastern R. Co., 298 US 492, 56 S Ct 841, 80 L Ed 1300 (1936). After the Elgin decision the Interstate Commerce Commission had entreated Congress to overrule that decision by legislation. As in the case before us, the Commission had caused a bill to be introduced for that purpose and Congress had refused to enact the requested legislation. In the South Buffalo case the United States asked the court, in effect, to reject its former interpretation of the statute. The court responded:

*817“It is the Government’s contention that the Elgin decision misconstrued the Act, misunderstood its legislative history and misapplied the Court’s own prior decisions. It is not necessary in the view we take of the case to decide to what extent, if any, these contentions are correct. It is enough to say that if the Elgin case were before us as a case of first impression, its doctrine might not now be approved. But we do not write on a clean slate. What the Court has written before is but one of a series of events, which convinces us that its overruling or modification should be left to Congress. As the Court held on our last decision day, when the questions are of statutory construction, not of constitutional import, Congress can rectify our mistake, if such it was, or change its policy at any time, and in these circumstances reversal is not readily to be made. Massachusetts v. United States, 333 U.S. 611, decided April 19,1948. Moreover, in this case, unlike the cited one, Congress has considered the alleged mistake and decided not to change it. ” (Emphasis added.)

333 US at 774-775, 68 S Ct at 870, 92 L Ed at 1081.

The similarity in the situation presented by the case at bar and that above described is striking. The same rule should apply.

Another court has recently come to a like result in a similar case, stating:

“The doctrine of stare decisis, weighty in any context, is especially so in matters of statutory construction. For in such cases Congress may cure any error made by the courts. Until it does, the bar and the public are justified in expecting the courts, except in the most egregious cases, neither to depart from previous interpretations of statutes, nor to give them a grudging application.”

Cottrell v. C.I.R., 628 F2d 1127, 1131 (8th Cir 1980) (footnote omitted).

Quite simply, the following situation obtains in the case at bar. This court interpreted the statute in Scharf. A concerned agency of the Executive Department sought to overturn that interpretation at the very next regular session of the legislature. The Legislative Department refused to amend the statute as requested. Now, due to a change in the personnel of this court, a new majority refuses to follow the rule of stare decisis in just the kind of case to which the rule is singularly pertinent, and the Judicial Department *818makes the change which the department of government charged with formulating statutory law has considered and refused to make. That course of action is both unwise and unseemly.

dissenting.

Less than two years ago, this court decided that when police officers unlawfully deny a driver’s request to telephone an attorney before choosing whether or not to submit to a breath test for intoxication, the results of such a test have not been obtained lawfully and cannot be used to convict the person of driving under the influence of intoxicants. State v. Scharf, 288 Or 451, 605 P2d 690 (1980). This was a decision under Oregon law, uncomplicated by constitutional problems. The decision could be changed by law, assuming the change to be otherwise constitutional. Today, this recent decision is overturned by a differently constituted plurality and one concurring opinion, although there has been no change in the law. Because the law remains what it was when Scharf was decided, because the plurality opinion needlessly enmeshes the law in serious constitutional difficulties, and because it leaves the people of Oregon with a legal right without an evident legal remedy, I dissent.

The court’s decision in State v. Scharf rested on two propositions. The first was that police officers do not have authority to prevent a person in custody from using an available telephone to call a lawyer (or for that matter, someone other than a lawyer), beyond what may be justified by the “immediate necessities of the arrest and the circumstances of custody itself.” 288 Or at 455. The court previously said as much in State v. Haynes, 288 Or 59, 602 P2d 272 (1979).1 It does not depart from that proposition *819today. The court reaffirms that it is unlawful for an officer to tell an arrested driver that he or she cannot make a telephone call until the driver first decides whether or not to submit to the breath test, as happened in this case.

The second proposition in State v. Scharf was that the unlawful prevention of a telephone call for legal advice, followed by the driver’s uncounseled submission to the breath test, vitiates the use of the test results against the driver in a criminal prosecution. This result followed because the “implied consent” law, ORS 487.805, does not provide for the involuntary administration of a breath test to a protesting suspect. To the contrary, it contemplates that the arrested driver may refuse to take the test at the price of a 120-day suspension of his or her driver’s license. ORS 482.540(1). Moreover, this sanction depends on a prior explanation to the driver of the consequences of refusing the breath tést, to be established in an adjudication in which the adequacy of the explanation can be contested. ORS 487.805(2), ORS 482.550. The driver who is “requested,” ORS 487.805(2), not commanded, to submit to the test therefore has a genuine choice with potentially decisive importance to his liberty. When a driver seeks legal advice before making that choice and is unlawfully prevented from doing so, there is an improper interference with the driver’s choice to take or refuse the breath test. The court concluded that the use of a test resulting from such improper interference to obtain a criminal conviction does not square with the statutory scheme that allows an informed driver to refuse the test at the price of a license suspension.

This conclusion in Scharf made it unnecessary to reach any constitutional question, at least unless and until the legislature might change the law. The legislature has not done so, as Justice Lent’s dissent points out. Nevertheless, the new plurality now chooses to plunge the court into needless constitutional speculations that ultimately lead it to a wholly incongruous conclusion.

Justice Tanzer’s opinion begins with a history of the so-called “implied consent statutes,” such as ORS 487.805, and its relationship to certain Supreme Court decisions. The sole object of this lengthy review is to argue *820that, although the statute directs the police to “request” a suspected driver to take the breath test, not to order or command the driver to do so, and although the statute permits the driver to refuse, nevertheless it does not leave him or her any choice. “Consent” and “refusal” are said not to be “antonyms.” With respect, that bit of verbal legerdemain flies in the face of a reality. Whether a refusal to submit to a breath test is a “refusal,” ORS 487.805(2), or a withdrawal of a fictitious “consent,” the effect is the same. The driver may choose not to submit to the test, and the test is not administered. The reason is not the driver’s use of any “physical power” to resist (291 Or at 793). To the contrary, the legislature chose to recognize the option to refuse in the statute itself.

The question therefore is not what effect a wrongful refusal to let an arrested person make a telephone call would have if the officer were authorized to enforce a breath test or other examination in any event. But he has no such authority. An officer who sought to coerce submission to a breath test over the protests of a resisting suspect plainly would not be following this statute.2 Even the *821sanction of license suspension can be imposed only if the officer first explains this consequence of persisting in the refusal. Obviously the objective of the “implied consent” laws is use of the test rather than refusal and license suspension; that hardly needs lengthy demonstration. Nonetheless, the statute preserves the choice of refusal. To call this provision a “threat” cannot hide the fact that it openly and candidly leaves a genuine choice to a driver who is far more threatened by a criminal conviction on the evidence of a breath test.

Part II of the opinion discusses a variety of constitutional limitations on the state’s authority to administer the breath test, which become relevant only because the new majority is determined to overturn State v. Scharf. In 11(C), the plurality launches into an essay on the political philosophy of liberty and the “Social Compact” in order to show that defendant’s right to call his lawyer derives from the 14th amendment of the United States Constitution. Whatever its merits as rhetoric, this section falls short as serious constitutional analysis.

First, invocation of the federal 14th amendment for an arrested person’s right not to be held incommunicado suggests that Oregon law would deny such a right if it were not restrained by higher federal authority. That is simply false, State v. Scharf, supra, State v. Haynes, supra, and it is an unwarranted affront to the law of this state. Moreover, in the plurality’s own terms, if there was such a thing as a “Social Compact,” it found its expression when Oregon Recame a state in 1859,3 well before the Civil War and the post-war adoption of the 14th amendment in 1868. It is wholly inconsistent with that theory to imply that Oregon law would allow police to hold arrested persons incommunicado if there had been no Civil War and no 14th *822amendment to impose a “Social Compact” on us. Quite apart from such a theory, however, this court repeatedly has held that it is not necessary to turn to restraints imposed on this state by federal law in order to protect the rights of Oregon citizens that their own law preserves in any event.4 If, indeed the conclusion that officers may not deny outside communication to one taken into police custody needed a constitutional source, it is most pertinently implicit in the constitutional guarantee of habeas corpus. Or Const Art I, § 23.5

Second, if the plurality is set upon its gratuitous essay into federal constitutional law, it should take its undertaking seriously. In questions of federal law, such as the 14th amendment, we follow the decisions of the United States Supreme Court, not our own views, as is constantly shown in the court’s treatment of fourth amendment claims. See, e.g., State *823v. Brown, 291 Or 642, 634 P2d 212 (1981). But the plurality makes no effort to determine what the Supreme Court has held or said bearing on the issue before us. Whatever the Supreme Court might hold in such a case, one thing is certain: Neither that court nor any court familiar with federal constitutional law would cite as the source of its “analysis” a quotation from a single justice dissenting from a dismissal of an appeal in a birth control case.6 The plurality’s quotation merely demonstrates the pitfalls of transferring large generalities from one area of law into a totally different one. I do not say that the Supreme Court would not hold that the 14th amendment protects an arrested person’s right to communicate with the “outside” world; I hope that the Court would so hold. But surely there are more relevant sources for understanding a person’s liberty not to be imprisoned incommunicado than to compare it with freedom to use contraceptives or to have an abortion. That is not a serious attempt to discern and to follow the federal law.

Third, after finding a violation of due process of law, the plurality opinion incongruously holds that a conviction resulting from that violation nevertheless will not deprive Mr. Newton of liberty without due process of law. The opinion finds that it was a violation of due process to deny Newton his requested opportunity to telephone for legal advice. As I have said, this constitutional holding is unnecessary, though it may be correct. The opinion then asserts that there is no proof of a “causal connection” between this denial and Newton’s submission to the breath test, because the record does not show that he would have reached his lawyer or what advice the lawyer would have given him.

*824The second part of that assertion is disingenuous; a court should assume that counsel would advise a client not to take the breath test unless he is certain of a negative result. Contrary to the plurality opinion, when the impropriety consists of interfering with a person’s access to counsel, we do not invite either the police or trial courts to “speculate” whether counsel might have been unavailable or might have advised cooperation with the police. Cf. State v. Haynes, supra. If the state wishes to assert that improper interference with access to counsel was harmless for such a reason, it must shoulder the burden of demonstrating that fact. But the fact is that in this case, as in Scharf, the officers did not prevent the telephone call to a lawyer because they thought it would be futile or immaterial to their investigation. To the contrary, they followed an official policy of preventing such calls precisely in order to forestall advice to decline the test. When the state adopts and enforces a policy of denying access to counsel for this very purpose, it is reasonable to assume that the denial has served that purpose unless the opposite is shown. It should also be obvious that the denial is designed to affect and does affect a critical stage in a criminal investigation, often the decisive stage.7

*825The breath test evidence in this case was obtained by deliberate, official interference with Mr. Newton’s efforts to call a lawyer, thus obtained unlawfully and, in the plurality opinion, contrary to due process of law. Nevertheless, that opinion maintains that a conviction based on this violation of due process would not be a violation of due process. On its face such a position is an anomaly. The opinion attempts to explain away the anomaly by a further disquisition on “competing policy considerations” and an “empirical and pragmatic approach,” concluding that due process does not need to be enforced in this case because it is probably not needed in order to deter future official violations. The explanation does not resolve the obvious contradiction, it only deepens it. The person whose liberty is not to be taken without due process of law is the present defendant, Kenneth Ray Newton. The person who, according to the plurality opinion, was deprived of access to counsel without due process of law is the present defendant, Mr. Newton. It is Mr. Newton whom the state intends to convict upon a process of which a crucial part was not “due process.” Whether the total process is or is not due process with respect to Newton cannot logically depend on this court’s guess as to the future policies of the state police toward other motorists. That kind of “pragmatism” has recently become a fashionable way to “explain,” and perhaps to undermine, the Weeks8 doctrine in federal search and seizure cases, as Justice Tanzer says, but I believe it has no proper place in assuring each individual brought before a court of due process of law. The state and federal constitutions do not make the individual person a mere instrument for judicial “policy considerations,” to be protected when a court thinks this will usefully influence official conduct and not protected when a court thinks that it will not. They do not invite courts to manipulate due process for the individual only as a means to some other end. Such an approach makes a mockery of a guarantee that government shall not deprive any person of life, liberty, or property without due process of law. US Const Amend V, Amend XIV.

Finally, the prevailing opinions in this case maintain a discreet silence on the question what remedy is available *826to Mr. Newton, or to others who despite the Court’s hopes may find themselves in his position. The truth is that the only relevant remedy for what the Court agrees was an unlawful, perhaps even unconstitutional, procedure is a trial untainted by that procedure. An arrested person whose request to telephone for help or counsel is wrongfully refused ordinarily is not a victim of physical or psychological mistreatment. His or her interest is not in punishing an officer who may only have followed erroneous instructions and otherwise have been wholly inoffensive. Nor is it in being made whole for a financial injury. The interest that is injured by the refusal is the arrested person’s interest in his or her legal rights when in police custody, and the injury directly threatened by the wrongful refusal is loss of liberty. One may ask what remedy the Court expects the next person to seek who learns after the fact that an officer’s refusal of a request to call counsel was unlawful. The one remedy that addresses the very interest which the telephone call could protect is to remove the fruits of the unlawful refusal from the prosecution. If that is denied, no similarly relevant remedy is evident, and the Court mentions none.

To sum up: Oregon law does not provide for subjecting an arrested motorist to a breath test over the motorist’s objection. The law provides that the motorist may refuse the test, with legal consequences for his or her license to drive. When an arrested motorist requests to use an available telephone to obtain advice whether to take or refuse the test, the request may not lawfully be denied, as the Court reaffirms in this case.

Given these premises, State v. Scharf held that when such a request to telephone one’s lawyer for advice was wrongfully denied, breath test results thereafter obtained from the arrested person could not be used in prosecuting that person. This inference drawn from the statutory recognition of the arrested person’s choice to refuse the test was not unanimous, but the disagreement remained a simple and straightforward question of Oregon law which could be addressed by the legislature if it wished. The legislature did not change the law, although it reexamined and amended the statutes governing driving under the influence of intoxicants.

*827Instead we now have a long and tortured opinion which seeks to explain that the reason why Oregon lets arrested persons use the telephone is that the federal 14th amendment compels Oregon to do so, that the refusal of permission to make the call in this case was a violation of due process under that amendment, and that nevertheless it is no violation of due process to prosecute this defendant on evidence obtained in apparent consequence of that refusal. Although that constitutional analysis does not command a majority of the court, its proposed shift of direction from Scharf is cause for regret.

We are dealing, not with convicted and imprisoned offenders, nor only with drivers suspected of intoxication. We are dealing with the rights of ordinary men and women who have been placed under arrest for an alleged traffic offense or any other reason and detained from proceeding toward their intended and expected destinations, whose unexplained disappearance can cause serious concern for others as well as inconvenience and trouble for themselves. A number of consequences no doubt are implicit in statutory authority to take a person into custody, but authority to prevent him or her from communicating with the outside world, particularly from seeking to obtain legal counsel, is not one of them.

Whether motorists are left the choice to refuse breath tests for intoxication perhaps is not of fundamental importance. An arrested person’s access to outside advice, particularly legal advice on a critical choice while in custody for potential prosecution, is of fundamental importance. A view that interference with such access would not be unauthorized in Oregon but for federally imposed restraints carries ominous implications beyond the details of the “implied consent” statutes. I would adhere to our previous decision under the same statutes until they are changed.

In this prosecution for driving under the influence of intoxicants, ORS 487.540, the district court suppressed evidence of a breathalyzer test of the defendant’s blood alcohol content and the state appealed. The Court of Appeals affirmed on the basis of our decision in State v. Scharf, 288 Or 451, 605 P2d 690, rehearing den 288 Or 621 (1980).

In State v. Scharf this court divided on whether a denial of access to counsel violated Oregon’s implied consent statute and required exclusion. A majority held that ORS 487.805 as a matter of legislative policy, assured to an arrested person a “voluntary and informed choice,” 288 Or at 458, whether to take a breathalyzer test which was violated by an unauthorized police refusal to allow defendant to call a lawyer for advice as to whether to consent. Three judges dissented on the ground that the legislative intent of ORS 487.805 was not to assure a voluntary and informed choice. We allowed review to re-examine our statutory determination.

FACTS

The defendant was arrested for driving under the influence of intoxicants at about 9 p.m. The validity of the arrest is not challenged. He was advised of his rights to silence and counsel. After the officer drove defendant home to leave his automobile keys and other property with his wife, defendant was taken to the county jail. At about 11 p.m. he was requested to take a breathalyzer test. The defendant then requested the opportunity to talk with a lawyer before the breath test was administered. A telephone was available in the jail at the time of the request. The officer administering the breathalyzer quoted from the standard Oregon State Police informational form and advised defendant that “regardless of any information you may have received before this request, the fact is you are not entitled to have an attorney present at this breath test and any request for delay on this ground will constitute a refusal.”1 Defendant then submitted to the breathlyzer test and the results indicated a blood alcohol *791content of. 10 percent. The district court suppressed that evidence on the basis that defendant was entitled to a telephone call to consult with a lawyer before taking the test.

I. THE IMPLIED CONSENT STATUTE

A threshold difficulty in discerning the legislative intent embodied in ORS 487.8052 is not simply that the concept of implied consent is a “statutory fiction,” Scharf, 288 Or at 457, but that the fiction appears to be theoretically contradictory. An enigma appears to be at the heart of the law: If, under subsection (1), a driver has impliedly consented to a breath test which, under subsection (2), he may nevertheless refuse, then “the licensee-driver has not impliedly consented to anything.” Lerblance, Implied Consent to Intoxication Tests: A Flawed Concept, 53 St Johns L Rev 39, 49 n. 36 (1978).

*792The contradiction disappears, however, when it is realized that the words “consent” and “refusal” are not used as antonyms, because they are not used in the same sense. “Consent” describes a legal act; “refusal” describes a physical reality. By implying consent, the statute removes the right of a licensed driver to lawfully refuse, but it cannot remove his or her physical power to refuse. As another court put it:

“The obvious reason for acquiescence in the refusal of such a test by a person who as a matter of law is ‘deemed to have given his consent’ is to avoid the violence which would often attend forcible tests upon recalcitrant inebriates. * * *

*793“* * * It is firmly established that a drunken driver has no right to resist or refuse such a test [citations]. It is simply because such a person has the physical power to make the test impractical, and dangerous to himself and those charged with administering it, that it is excused upon an indication of his unwillingness. * * *” Bush v. Bright, 264 Cal App 2d 788, 790, 792, 71 Cal Rptr 123 at 124, 125 (1968) (original emphasis).

Thus refusal as contemplated by the statute is something other than withholding of consent because consent is legally implied. It is a refusal to comply with-the consent which has already been given as a condition of a license to drive. The purpose of a warning of license suspension following a refusal is to overcome an unsanctioned refusal by threat instead of force. It is not to reinstate a right to choice, let alone a voluntary and informed choice, but rather to nonforcibly enforce the driver’s previous implied consent.

The history of the . implied consent law confirms the preceding description of the nature of consent and refusal as those terms are used in implied consent statutes. The law is designed to overcome the possibility of physical resistance, despite legal consent, without resort to physical compulsion. Tracing the history of the statute, we find that it is intertwined with constitutional caselaw, uniform legislation, and federal funding. The conclusion is clear that the concept of implied consent is rooted in a misconception of the law of due process which has become legislatively perpetuated even though the constitutional underpinnings have long since been superseded.

Implied consent statutes were initially a reaction to caselaw. The seminal case is Rochin v. California, 342 US 165, 72 S Ct 205, 96 L Ed 183 (1952). There, the police, armed only with information which was less than probable cause to believe that the defendant was selling narcotics, surreptitiously entered his home and forcibly entered his room without a warrant. Defendant, seated on his bed next to his reclining wife, seized two capsules from his bedstand and placed them in his mouth. Three officers struggled unsuccessfully to extract the capsules from his mouth. Defendant was handcuffed and taken to a hospital where, at police direction and against defendant’s will, a doctor *794forced an emetic solution through a tube into defendant’s stomach, causing him to vomit. In the vomit were two capsules of morphine.

The court held that the course of conduct by which the police obtained the evidence offended “those canons of decency and fairness which express the notions of justice of English-speaking peoples” embodied in the due process clause of the Fourteenth Amendment of the United States Constitution, 342 US at 169, because

“ * * * This is conduct that shocks the conscience. Illegally breaking into the privacy of the petitioner, the struggle to open his mouth and remove what was there, the forcible extraction of his stomach’s contents — this course of proceeding by agents of government to obtain evidence is bound to offend even hardened sensibilities. They are methods too close to the rack and the screw to permit of constitutional differentiation.” 342 US at 172.

Although later cases made it clear that Rochin was based upon the totality of the police conduct, just as the opinion said, the decision was thought at the time to bar the forcible invasion of the body for seizure of evidence as violative of due process. That being so, there was concern that the Rochin theory invalidated state statutes authorizing the extraction from the body of blood, urine or other substances for testing of alcohol content. See Comment, Constitutional Law — Validity of New York Statute Setting out Motorists’ Implied Consent to Chemical Tests for Intoxication, 51 Mich L Rev 1195 (1953).

One year later, the New York legislature acted in response to the Rochin decision by enacting the prototype implied consent statute. See Note, Driving While Intoxicated and the Right to Counsel: The Case Against Implied Consent, 58 Tex L Rev 935,938 (1980). Unlike former statutes, the new statute did not authorize physical compulsion to overcome resistance to a demand for a blood sample, a practice thought to have been barred by Rochin. Rather, the statute authorized police to overcome refusal by the threat of an adverse consequence, i.e., license suspension. Nothing suggests a legislative intention to offer the driver a free choice or to create the statutory equivalent of a consent search; the objective was to create a means of *795non-physical coercion in order to obtain chemical samples with which to convict drunk drivers. Implied consent statutes modeled on the New York statute say only that the test “shall be administered” on request of the officer. Only if there is a refusal to submit are further procedures required.

Prior to 1965, Oregon law authorized blood, breath, or urine tests only upon the actual consent of a person arrested for driving under the influence of intoxicating liquor.3 This was changed in 1965 when Oregon adopted the implied consent act essentially in its present form. Its proponent explained:

“The question of consent remains the crux of the problem today. Although we have a chemical test law to deal with the problem of the drinking driver, it is not fully utilized due to the difficulties in obtaining consent [i.e. submission]. Refusals to take a chemical test run as high as 70. in some jurisdictions. On an average, about 2/3 refuse to take a test.” Testimony of Warne Nunn, Chairman, Oregon Traffic Safety Commission, representing Governor Mark Hatfield, 1-28-65, House Judiciary Committee.

The proponent continued to explain that consent would continue to be required for blood tests. For breath testing, however, an arrested person need not consent because his consent was to be implied, but the test would not be administered if refused and if the refusal is maintained after advice of the consequences.4

*796Meanwhile, the constitutional assumptions arising from Rochin, upon which the implied consent act was based, had begun to erode. In 1957 a divided United States Supreme Court decided in Breithaupt v. Abram, 352 US 432, 77 S Ct 408, 1 L Ed 2d 448 (1957), that a blood sample taken at the hospital from an unconscious driver was constitutionally obtained even though the body had been invaded without consent. In particular, it held, there was no “coercion” or “brutality” as in Rochin. The court held that properly drawn blood tests had become commonplace and the subjection of a person to such a test did not shock the conscience. The evidence of testing was therefore not subject to exclusion. Four justices dissented, however, on the ground that the Rochin principle was offended by any invasion of the body whether or not physical force was used. The result of Breithaupt v. Abram was that Rochin was adhered to, but the narrowness of the holding and its implications was clarified.

In the same year that the broad constitutional implications of the Rochin doctrine were rejected by the United States Supreme Court, an important legislative event tended to lock those very implications into statutory law. In 1957 the Uniform Chemical Test for Intoxication Act was promulgated. See Handbook of the National Conference of Commissioners on Uniform State Laws and Proceedings of the Annual Conference 216, 220 (1957). In 1962 implied consent was adopted as part of the Uniform Vehicle Code, § 6-205.1 (1962).5 Thus legislation conceived to overcome a constitutional obstacle gained institutional momentum at the very time that the United States Supreme Court removed the obstacle.

Shortly after Oregon’s enactment of ORS 487.805 (then 483.634), the United States Supreme Court removed any lingering question that consent, whether implied or acutal, was necessary to overcome constitutional limitations upon mandatory blood testing. In Schmerber v. California, 384 US 757, 86 S Ct 1826, 16 L Ed 2d 908 (1966), the *797court held that, under proper circumstances, the state could constitutionally compel a person to submit to extraction of blood for testing to determine alcohol content. The Court upheld the use in a prosecution for driving under the influence of liquor of test results from a blood sample taken by a physician at the direction of a policeman from a conscious person over his objection. The Court stated, as it had in Breithaupt v. Abram, that the extraction did not offend the “sense of justice” referred to in Rochin, that there was no compelled self-incrimination under the Fifth Amendment because the evidence was not testimonial, and that the seizure was permissible under the Fourth Amendment because there was probable cause to believe evidence of crime was in defendant’s blood and that delay would likely destroy the evidence, and because the manner of extraction was reasonable. Again, four judges dissented. The Schmerber decision was consistent with preexisting Oregon law, see State v. Cram, supra, n 5, and neither court has withdrawn from that constitutional principle. After Schmerber, it has been observed:

“ * * * Paradoxically, as a result of Schmerber, states without ‘implied consent’ statutes can compel a motorist to undergo a chemical test, while in states with ‘implied consent’ statutes the motorist has the option to refuse the test.” Comment: Florida’s ‘Implied Consent’ Statutes: Chemical Tests for Intoxicated Drivers, 22 U Miami L Rev 698, 727 (1968).

In the same year as Schmerber, the legislative movement for implied consent gained another source of momentum. Congress adopted the Highway Safety Act of 1966, 23 USC 401-406. One year later, the Department of Transportation formulated standards under the Act, applicable to the states, including Highway Safety Standard No. 8, which required an implied consent enforcement scheme. See H.R. Doc. No. 138, 90th Cong., First Sess. 9,1967. All 50 states now have implied consent laws, almost half of them adopted since Schmerber. Although it has been suggested that post-Schmerber adoption of such laws reflects a legislative desire for greater protection of arrested drivers than is constitutionally required, see Hunvald & Zimring, Whatever Happened to Implied Consentí A Sounding, 33 Mo L Rev 323, 323-324 (1968), and Lerblance, supra at 47 n 22, no evidence or history is submitted to support that view.

*798Rather, one commentator has concluded in view ot this federal legislative history that:

“ * * * [I]mplied consent is a much a result of historical snowballing as of a considered choice by legislatures, * * *.” 58 Tex L Rev at 959.

Thus it is, he stated, that

“ * * * Although implied consent developed as a means of facilitating the use of chemical evidence, the statutes currently constitute the primary barriers to the use of that evidence * * Ibid.

Despite the decision in Schmerber, implied consent has not been repealed in Oregon. To do so would threaten the receipt of funds under the Highway Safety Act. There have been two amendments, however, which are significant as to legislative intent regarding consent and refusal. In 1975, as an ancillary to enactment of a revised traffic code, the license suspension period under ORS 482.540(1) for refusal to take a breath test was increased from 90 days to 4 months, Oregon Laws 1975, ch 451, § 146. The reason was that with the “decriminalization” of the traffic code, “people were going to refuse to take the test” unless the penalty for refusal was made more severe. Testimony of Chief Judge Herbert M. Schwab, 6-12-74, Interim Committee on Judiciary. In 1979, another adverse consequence of refusal was provided. The legislature enacted ORS 487.805(4), which allows evidence in subsequent legal proceedings of refusal to take the breath test. Both legislative changes were intended to compel submission rather than to promote freedom of choice.6

The preceding history does not support the conclusion in Scharf that ORS 487.805 embodies “the legislative decision that the breath test is to be administered only upon the arrested person’s voluntary and informed choice.” 288 or at 461. Rather, it confirms our earlier conclusion in State v. Fogle, 254 Or 268, 459 P2d 873 (1969), that the option of refusal under the statute does not imply a requirement of “voluntary submission”:

*799“* * * While the statute recognizes that a person may refuse to submit to the test, the legislature could hardly have contemplated that it was necessary that there be a completely knowing and understanding submission. If this were the case, the only people who could be tested would be those who were not sufficiently intoxicated to interfere with their mental processes.” 254 Or at 270.

Our description of legislative meaning in Fogle was consistent with the history of the statute. The very concept of implied consent in ORS 487.805(1) was intended to eliminate the right of choice and to recognize actual choice only in the sense of a forbearance of physical resistance. See Bush v. Bright, supra. Under ORS 487.805(2), the arrested driver must make the initial choice to submit or refuse without the benefit of any information from the police. If anything, this indicates a legislative intent to promote uninformed submission rather than informed choice. The police are obliged to give the arrested driver information only after he first “refuses the request of the police officer to submit to the chemical test” of his breath. Had the legislature been concerned “with assuring the arrested driver a voluntary and informed choice” as we stated in Scharf, 288 Or at 459, it surely would have required the police to advise the driver before, not after, the driver first chooses to submit or refuse. If the breath seizure were “the very stage that the legislature expressly required to be voluntary and informed” as we stated in Scharf, 288 Or at 460, the legislature would have required that the information be provided before, not after, drivers provide the evidence by submitting to the officer’s initial request.

Instead, ORS 487.805(2) provides that for drivers who comply with the officer’s initial request for a breath sample, the process is at an end even though the driver may never have been advised of anything. For drivers who refuse, the purpose of the statute is to provide an effective means short of physical force (in view of Bochin) to overcome the refusal. The nonphysical means consist of the statutory creation of a penalty (i.e., an extensive license suspension) and a warning from a police officer (as police officer, not as surrogate defense lawyer) that the driver’s license will be suspended if he or she continues to refuse. Thus the consent envisioned by the statute is to be implied *800and if submission is not forthcoming, it is to be coerced by fear of adverse consequences. Defendant’s request to call a lawyer is not material to statutory compliance.

In this case, defendant’s consent had alredy been impliedly given in the sense that he accepted testing as a condition of his license to drive. His submission was obtained in the statutory manner. Therefore, the implied consent statutes were complied with and they give no basis for suppression of evidence of the results of the chemical test of the defendant’s breath.7

II. CONSTITUTIONAL CLAIMS

Defendant has also relied throughout on the Oregon and United States Constitutions as bases for suppression. Although his arguments in defense of the order of suppression have centered on Scharf, he has expressly continued to assert his constitutional rights to freedom from unreasonable seizures and to counsel. Having held that ORS 487.805 was not violated, we now address defendant’s constitutional assertions.

A. Seizure

We begin by examining the constitutional propriety of the seizure of defendant’s breath sample. The Oregon State Police are charged with enforcement of Oregon’s traffic laws, ORS 181.040, and they are therefore authorized to take such actions as reasonably tend to accomplish that statutory responsibility and which are not prohibited by law. Obviously, obtaining evidence of traffic offenses is action which has a reasonable tendency to accomplish traffic law enforcement. In constitutional terms, the taking of physical evidence, e.g., defendant’s breath, is a seizure and unreasonable seizures are prohibited by the Fourth Amendment of the United States Constitution and Article I, section 9, of the Oregon Constitution. We must therefore decide whether these constitutional limitations on the authority to seize were transgressed in this case.

*801There was no warrant for the seizure. Therefore, if the seizure is constitutionally valid, it must come within an exception to the warrant requirement.

1. Consent

The warrant requirement may be excused if there is consent. By this, we mean actual consent. Defendant’s statutorily implied consent cannot excuse an otherwise unconstitutional seizure. If defendant’s submission may be regarded as a consent to seize, it was an informed consent in that defendant knew that he had the option of refusal, cf. State v. Flores, 280 Or 273, 570 P2d 965 (1977). The controlling principle is that effective consent must be voluntary. That is, the consenting person’s will may not be overborne. Schneckloth v. Bustamonte, 412 US 218, 225-226, 93 S Ct 2041, 36 L Ed 2d 854 (1973).8 Where a person’s consent to a seizure is solicited, and the person consents only after being warned that he will suffer a substantial penalty if he refuses, the resulting consent cannot be regarded as a free exercise of will. We therefore hold that defendant’s submission to the breath test was not a voluntary consent to seizure because it was coerced.

2. Probable Cause; Exigent Circumstances

The warrantless seizure was nevertheless not forbidden by the Fourth Amendment or Article I, section 9, because there was both probable cause to believe that defendant’s breath contained evidence of intoxicants and exigent circumstances making it likely that the evidence would dissipate if time were taken to obtain a warrant for its seizure. The validity of the arrest for driving under the influence of intoxicants is not challenged and blood alcohol content is a transitory condition. In such a case, neither constitution prohibits the police from compelling a person to submit to physical extraction of blood over the refusal of the person and the objection of his lawyer. Schmerber v. California and State v. Heintz, 286 Or 239, 594 P2d 385 *802(1979); accord re fingernail scrapings, Cupp v. Murphy, 412 US 291, 93 S Ct 2000, 36 L ed 2d 900 (1973). The assistance of counsel is not material to the validity of the seizure because defendant has no right to refuse and “he has no greater right because counsel erroneously advised him that he could assert it.” Schmerber, 384 US at 765-66. That is a dominant constitutional principle of this case: statutory considerations aside, the police could have lawfully compelled defendant to provide a breath sample for testing regardless of whether defendant called a lawyer and regardless of whether defendant had refused to submit to the seizure. Therefore, we conclude that the search and seizure provisions of neither constitution were violated.

B. Right to Counsel

Defendant’s right to counsel is material because his request to call a lawyer was discouraged by advice that delay therefor would be deemed a refusal, resulting in a license suspension. In effect, the discouragement of defendant’s request was a denial of it.

Neither Article I, section 9, nor the Fourth Amendment forbid compulsion; they allow compulsion when warranted or otherwise reasonable. The right of counsel does not apply under the Fourth Amendment as it does under the Fifth Amendment because a reasonable search and seizure need not be voluntary. Rather, as we said above, seizure may be compelled regardless of the action of an attorney. Schmerber v. California, State v. Heintz, Cupp v. Murphy, all supra. No right of counsel is implied by the right of freedom from unreasonable seizures. Therefore, the discouragement of defendant’s call to an attorney has no effect on the validity of the seizure in this regard.

It is also arguable that counsel is required because the breathalyzer request is a critical stage of the prosecution. See State v. Fitzsimmions, 93 Wash 2d 436, 610 P2d 893 (1980). The right to counsel attaches to certain evidence-gathering processes which are deemed “critical stages” of the prosecution as an extension of a defendant’s right to representation by counsel in court. Any pre-trial adversarial contact of the state and a defendant at which some benefit of counsel would be lost if counsel is not present, *803that is, at which the state’s case may be enhanced or the defense impaired due to the absence of counsel, may be considered a critical stage of the prosecution at which defendant has a right to the presence of counsel. Thus, under the Sixth Amendment, surreptitious questioning of an indicted defendant in the absence of counsel is deemed a denial of counsel at a critical stage of the prosecution. Massiah v. United States, 377 US 201, 84 S Ct 1199, 12 L Ed 2d 246 (1964). Similarly, a defendant is entitled to the presence of counsel at a post-indictment line-up because it is deemed a critical stage of the prosecution. United States v. Wade, 388 US 218, 87 S Ct 1926, 18 L Ed 2d 1149 (1967), Gilbert v. California, 388 US 263, 87 S Ct 1951, 18 L Ed 2d 1178 (1967). Evidence obtained at a critical stage at which the right to counsel has not been afforded or waived is subject to exclusion.

The breathalyzer test in this case cannot be characterized as a critical stage. Any evidence gathering, however early in the investigation or prosecution, may affect the ultimate adjudication and could therefore logically be regarded as a critical stage. A breathalyzer request and submission is such a process. Yet, no law requires the police to secure a lawyer for every person from whom evidence is sought at any time in the course of an investigation at the risk of having evidence suppressed if the person is ultimately charged. Not every evidence-gathering procedure is a critical stage. Obviously, the critical stage concept has a limit and that limit was set by the U. S. Supreme Court in Kirby v. Illinois, 406 US 682, 92 S Ct 1877, 32 L Ed 2d 411 (1972). The majority upheld evidence from a post-arrest, pre-indictment identification procedure. It held that an adversarial contact is a criticad stage in the prosecution only after the defendant is formally charged:

“The initiation of judicial criminal proceedings is far from a mere formalism. It is the starting point of our whole system of adversary criminal justice. For it is only then that the government has committed itself to prosecute, and only then that the adverse positions of government and defendant have solidified. It is then that a defendant finds himself faced with the prosecutorial forces of organized society, and immersed in the intricacies of substantive and procedural criminal law. It is this point, therefore, that marks the commencement of the ‘criminal prosecutions’ to *804which alone the explicit guarantees of the Sixth Amendment are applicable. [Citations omitted.]” 406 US at 689r690.