concurring and dissenting:

Although I concur in part II B of the opinion, I must dissent from part II A. If this case presented a matter of first impression, the majority’s position would not be unreasonable. However, the issue is not novel and we are not free to resolve it as the majority proposes.

It is important to keep in mind the narrow nature of the issue in this case. All agree that if there is a contract, then the arbitrator has jurisdiction. On the other hand, all agree that if there is no contract then the district court has jurisdiction. The dispute is thus limited to the question of whether an arbitrator or the district court will make the threshold determination of whether a contract has been formed.

The majority holds that before an arbitration clause in a contract can bind a party to arbitrate the party must be given an opportunity to dispute the “very existence” of the contract in district court. The majority’s view is the product of a powerful *1145 intuition. The majority strongly believes that “bootstrapping” should be avoided; a party should not be subjected to arbitration and excluded from court absent a prior judicial determination that the party has agreed to arbitration. See opinion at 1142 (citing AT & T Technologies, Inc. v. Communications Workers, 475 U.S. 643, 648 , 106 S.Ct. 1415, 1418 , 89 L.Ed.2d 648 (1986)). For “[ajrbitration is a matter of contract and a party cannot be required to submit any dispute which he has not agreed so to submit.” Id.

However, the concern that moves the majority is not the only policy consideration in this area of the law. There is a very strong federal policy in favor of arbitration. Moses H. Cone Memorial Hospital v. Mercury Constr. Corp., 460 U.S. 1, 24 , 103 S.Ct. 927, 941 , 74 L.Ed.2d 765 (1983). Indeed, “any doubts concerning the scope of arbitrable issues should be resolved in favor of arbitration....” Id.

Moreover, since so many contract disputes concern the “making” of the contract, rather than its voidability, the potential scope of contract arbitration will be severely limited by a rule that requires a judicial determination that a contract exists before the arbitration clause of such a contract can be effective. Consider a number of illustrations. First, the question whether there is consideration for exchanged promises or performance goes to the “making” of a contract. For if there is no consideration, no contract has been formed. See E. Farnsworth, Contracts 39-105 (1982). Second, the question whether a purported contract is actually a contract or is a mere preliminary negotiation also goes to the “making” of a contract. See id. at 113-122 (1982). Finally, consider the instant case, in which an agent is said to have lacked authority to bind her purported principal. Again, this goes to the “making” of the contract. Thus, if disputes that concern the “making” of contracts must be litigated rather than subjected to arbitration, arbitration’s potential as an alternative dispute resolution mechanism is substantially limited. As such, the majority’s policy concern about bootstrapping is counterbalanced by the contradictory federal policy that arbitration of contract disputes is to be encouraged.

The majority, based on its concern about bootstrapping, concludes that the district court should determine whether a contract has been “made.” Others, perhaps acting out of concern that the majority’s rule would unduly limit the potential of arbitration, have reached different conclusions. Three controlling authorities instruct that disputes about whether a contract has ever been “made” are to be decided by the arbitrator: Congressional statute, Supreme Court precedent, and Ninth Circuit precedent. We cannot legitimately ignore these authorities.

Congress expresses its view in the Federal Arbitration Act (“the Act”). Section 4 of the Act provides that federal courts who would have jurisdiction over the contract dispute but for the arbitration provision shall enter orders compelling arbitration “upon being satisfied that the making of the agreement for arbitration ... is not in issue_”. 9 U.S.C. § 4 (emphasis added).

The Supreme Court interpreted § 4 in Prima Paint v. Flood & Conklin Mfg. Co., 388 U.S. 395 , 87 S.Ct. 1801 , 18 L.Ed.2d 1270 (1967). The Court held that a claim of fraud in the inducement to contract did not support district court jurisdiction under § 4. Rather, the arbitrator should deal with the claim. The Court reasoned that, in light of § 4’s language, for a claim of fraud in the inducement to make the dispute non-arbitrable the claim must specifically relate to the arbitration clause itself, instead of merely impugning the contract generally. Id. at 402-04, 87 S.Ct. at 1805-06 . The Court said:

*1146 Id. at 402 , 87 S.Ct. at 1805 . 1

In dissent, Justice Black asserted the very view now adopted by the majority in this case, that the statute cannot be interpreted so as to allow “bootstrapping.” Justice Black argued that if a contract is fraudulently induced it is not enforceable, and any arbitration provision within it is therefore ineffective. See id. at 407, 425 , 87 S.Ct. at 1807, 1817 (Black, J., dissenting). This view was necessarily rejected by the Prima Paint Court.

The majority seeks to distinguish Prima Paint on the grounds that the present case involves the question whether any agreement was ever reached, while Prima Paint involved the question whether an agreement was voidable due to fraud in the inducement to contract. As stated by the district court below:

In either case, no independent challenge is made to the arbitration clause itself. In both cases the contract is unenforceable.

Moreover, the Ninth Circuit has already considered whether Prima Paint extends to cases that turn on whether a contract has been “made.” Teledyne, Inc. v. Kone Corp., 892 F.2d 1404 (9th Cir.1990). 2 We concluded that Prima Paint does apply, so that issues relating to the making of contracts are subject to arbitration. Id. at 1410-11. Despite the majority’s intimations to the contrary, that Prima Paint formed the basis for Teledyne could not be clearer. We said:

Id. at 1410 (citations omitted). It is thus clear that in Teledyne this court held that, absent a specific challenge to the arbitration clause itself, a claim that a contract has never been made is arbitrable.

The majority attempts to distinguish Tel-edyne in two ways, neither of which is convincing. First, contrary to the majority’s claim, it is evident from the opinion that the Teledyne court did not import into § 4 an equitable rule that the party suing on a contract must submit to arbitration. And the Act contains no such requirement. Further, the equities of such a rule are far from apparent. For the very party who denied the existence of the contract was able to force the other party into arbitration.

The majority’s second ground for distinguishing Teledyne is no more convincing. *1147 The claim, incredibly, is that there was no dispute in Teledyne as to whether a contract had been “made,” because the parties admitted that they had signed something. One party said this something was a part of preliminary negotiations, and the other party said this something was a contract.

Of course there was a dispute in Tele-dyne as to whether a contract had been formed. If the signed paper was a mere negotiating instrument, then the contract that would have given a basis for arbitration never existed. For a mere negotiating instrument has no more effect than a paper signed by one without authority to bind his principal. Would a blurred scribble on a dinner napkin also satisfy the majority that a contract had unquestionably been executed? Thus, the issue in Teledyne, as in this case, was whether the contract, of which the arbitration clause was a part, had been formed. And Teledyne holds that absent a specific challenge to the arbitration clause this issue is for the arbitrator.

In short, the majority’s stance is simply untenable. To evade the constraints of the three sources of authority just discussed the majority would import into 9 U.S.C. § 4 a number of novel distinctions. The majority would first distinguish between “void” and “voidable” contracts, thereby severely limiting the types of contract disputes that may be brought before an arbitrator. The majority would next distinguish between eases in which the plaintiff seeks arbitration and cases in which the defendant seeks arbitration. The majority’s final point is that arbitration is appropriate if any thing has been signed, so long as not just any body signed it. Rather than transform this area of the law into a morass of questionable distinctions, I would follow the commands of the statute and controlling authorities. I therefore respectfully dissent.

. The majority relies on a number of labor cases. See, e.g., AT & T Technologies, Inc. v. Communications Workers, 475 U.S. 643, 648 , 106 S.Ct. 1415, 1418 , 89 L.Ed.2d 648 (1986). These cases are not predicated on the Federal Arbitration Act, on which both Prima Paint , discussed infra, and the instant case are based, but on the Labor Management Relations Act. Id. at 646 & n. 4, 106 S.Ct. at 1417 & n. 4; Camping Constr. Co. v. District Council of Iron Workers, 915 F.2d 1333 (9th Cir.1990). Prima Paint's construction of the Federal Arbitration Act thus controls this case, rather than any general principles enunciated in these labor cases.

. The circuits are currently split on the issue in this case. Compare Par-Knit Mills, Inc. v. Stockbridge Fabrics Co., 636 F.2d 51 (3d Cir.1980) (not arbitrable) with Unionmutual Stock Life Ins. Co. v. Beneficial Life Ins. Co., 774 F.2d 524, 528-29 (1st Cir.1985) (arbitrable unless there is a specific challenge to the arbitration clause). The cases from other circuits that the majority cites must of course yield to the Ninth Circuit precedent of Teledyne.