(concurring). I agree with the result reached by the majority in this case but cannot agree with the analysis by which it reaches that result. The majority’s attack on People v Alvin Johnson, 396 Mich 424; 240 NW2d 729 (1976), the Michigan case which construes United States v Dinitz, 424 US 600; 96 S Ct 1075; 47 L Ed 2d 267 (1976), is unwarranted.
I believe Alvin Johnson correctly interprets Dinitz to require something more than a defendant’s attorney’s acquiescence to a waiver of double jeopardy. Dinitz states,
"The important consideration, for purposes of the Double Jeopardy Clause, is that the defendant retain *300primary control over the course to be followed in the event of such error.” 424 US at 609.
Alvin Johnson, interpreting that language, holds:
"[T]he defendant must therefore do something positively in order to indicate he or she is exercising that primary control.” 396 Mich at 432-433.
The guarantee against double jeopardy is a fundamental right. Benton v Maryland, 395 US 784; 89 S Ct 2056; 23 L Ed 2d 707 (1969), People v Cooper, 398 Mich 450; 247 NW2d 866 (1976). As Alvin Johnson correctly observed, it is a right which defendant himself or herself must waive.
Arguably, the "knowing, voluntary and intelligent” waiver concept is not applicable in Michigan because of the footnote in Dinitz which states that the Supreme Court rejects the concept. I can only note, however, that after Dinitz and after Alvin Johnson, our Supreme Court, in People v Cooper, supra, chose to use a waiver analysis to permit appellate consideration of a double jeopardy claim. The majority calls this analysis mistaken because of the Dinitz footnote. But it may also be viewed as a reflection of Michigan law in an area in which that law is beginning to vary from its Federal counterpart. Such action by our Supreme Court is not without precedent. See, for example, People v Beavers, 393 Mich 554; 227 NW2d 511 (1975).
But whether the analysis be traditional waiver or not, Alvin Johnson and Dinitz do require that defendant have primary control over the course to be followed. I believe that primary control must reside in the defendant himself or herself, or, as Alvin Johnson says, personally. The record must disclose a personal manifestation by defendant *301that she or he consents to a waiver of the protection against double jeopardy. It is not necessary, though it is commendable, that the record contain a statement by the defendant; but it must be demonstrated that counsel has at least spoken with his or her client about the constitutional protection and that after such discussion defendant has made the decision to consent.
The passages quoted at length by the majority are equally supportive of our respective positions. Only the supplied emphasis is different.1 Indeed, that Alvin Johnson uses the term "defense counsel” is not inconsistent with the view that a de*302fendant’s attorney can, in a sense, waive a defendant’s right, as long as counsel affirmatively indicates on the record that the defendant has had the primary control in the double jeopardy decision.
The majority’s references to "settled precedent,” i.e., its references to People v Gardner, 62 Mich 307; 29 NW 19 (1886), People v White, 68 Mich 648; 37 NW 34 (1888), People v Carlton Brown, 23 Mich App 528; 179 NW2d 58 (1970), I find both unenlightening and unpersuasive.
Alvin Johnson represents a discernible modification of prior case law. "Settled precedent” is, therefore, of limited value.' Even so, I do not agree that Alvin Johnson disregards settled precedent. The majority quotes People v Gardner, supra, on this point. Gardner states:
"But to have this operation and effect such discharge must have been made without the consent, express or implied, of the respondent.” 62 Mich at 311. (Emphasis supplied by majority.)
Alvin Johnson, rather than disregarding the Gardner language, actually expands upon the concept of "implied” used there, by requiring that an implied consent, i.e., one made by counsel for his or her client, be the result of a procedure in which defendant had primary control. The only effective way to ensure that defendant has primary control is for defendant himself or herself to consent on the record or for defendant’s counsel to establish on the record that the decision to consent was the outcome of a process over which defendant had primary control.
From the record in this case I am led to believe, assuming but not deciding the retroactivity of Alvin Johnson, that defendant had the requisite primary control. Accordingly, I concur.
To demonstrate that a change in supplied emphasis sustains an interpretation contrary to that urged by the majority, consider the following:
"In a very recent case in which defense counsel himself requested a mistrial, the United States Supreme Court has addre'ssed this point as follows:
" 'The important consideration, for purposes of the Double Jeopardy Clause, is that the defendant retains primary control over the course to be followed * * * ’, United States v Dinitz, 424 US 600, 609; 96 S Ct 1075, 1081; 47 L Ed 2d 267, 275 (1976), and the defendant must therefore do something positively in order to indicate he or she is exercising that primary control. Further,
" 'The circumstance that it is defense counsel who initiates the court’s inquiry into a matter which ultimately results in an order of mistrial does not ipso facto transform counsel’s expression of concern into an implied consent to such drastic ruling.’ People v Compton, 6 Cal 3d 55, 62; 98 Cal Rptr 217, 221; 490 P2d 537, 541 (1971).
" '[A]n appellate court’s assessment of which side benefited from the mistrial ruling does not adequately satisfy the policies underpinning the double jeopardy provision’, United States v Jorn, 400 US 470, 483; 91 S Ct 547; 27 L Ed 2d 543 (1971). Thus, in the absence of bright-line rules deliberately eschewed by the United States Supreme Court, 400 US 470, 486, we must Srst look to whether in the exercise of that control of the course of his own trial, highlighted by Dinitz, defendant Alvin Johnson approved termination of the proceedings.
"It is not difficult to require a trial court to inquire whether defendant consents. Therefore, in the absence of an affirmative showing on the record, this Court will not presume to find such consent.
"There was no such affirmative showing in this case. At best, defense counsel may be said to have been silent. At worst, he did not protest, but he did not assent. Therefore, we find that defendant did not personally consent to the end of the first trial. However, the finding of manifest necessity would still make the second trial permissible.” 396 Mich at 432-433. (Emphasis supplied.)