OPINION

MOSK, J.

Azusa Municipal Code section 8.52.060 (hereinafter the ordinance) provides that "No person shall practice or profess to practice or engage in the business or art of astrology, augury, card or tea reading, cartomancy, clairvoyance, crystalgazing, divination, fortune telling, hypnotism, magic, mediumship, necromancy, palmistry, phrenology, prophecy, or spiritual reading, or any similar business or art, who either solicits or receives a gift or fee or other consideration for such practice, or where admission is charged for such practice." [1]

As will appear, we conclude the ordinance is unconstitutional because it unduly burdens rights guaranteed by article I, section 2, of the California Constitution.

There is no material dispute regarding the facts. Plaintiff Spiritual Psychic Science Church of Truth, Inc. (the Church) was incorporated in 1976 as a tax-exempt nonprofit corporation. Plaintiff Fatima Stevens acted as its duly ordained minister. In June 1979 the Church was granted a business license by defendant City of Azusa (the City). The license, issued after Stevens had assured the city attorney that it would be used only to conduct religious workshops and counselling and not to practice fortunetelling, listed as the Church's business "spiritual, Palm Reading." Stevens had been informed by the attorney that fortunetelling and related activities, when practiced for consideration, would be unlawful under the ordinance, but that if these activities were pursued as a religious ritual the ordinance would not be violated. Stevens declared she had to charge a fee for telling fortunes because that was the source of her livelihood.

In August 1979 Stevens placed an advertisement in a local newspaper announcing that fortunes were being told at her place of business. There is some uncertainty in the record as to whether the Church's business license expired thereafter and was not renewed by the City or whether the City cancelled the license outright. In either event, the Church lost its license. In addition, Stevens was threatened with the possibility of prosecution under the ordinance. Therefore, the Church and Stevens filed suit against the City in 1981 to obtain an injunction enjoining defendant "from conducting religious spiritual services on said property [ sic ]." [2] They contended the ordinance was invalid on various constitutional grounds. The trial court denied an application for a preliminary injunction, and plaintiffs appeal. (Code Civ. Proc., § 904.1, subd. (f).) Before us plaintiffs argue that the ordinance violates their rights of free speech. (Cal. Const., art. I, § 2.)

The City contends the ordinance is a valid regulation of a commercial activity because it prohibits fortunetelling only for consideration. The City relies on Azusa Municipal Code section 5.21.010 et seq., which permit solicitation of general contributions for religious purposes, and specifically authorize solicitation of gifts from legitimate church members at assemblies, services or otherwise. The City asserts that Stevens may practice fortunetelling and may solicit funds for her church's religious activities; what she cannot do is engage in the business of telling fortunes for money or other consideration. Thus it is contended that the ordinance regulates not speech but commercial activity, and does so validly.

As support for its position, the City cites In re Bartha (1976) 63 Cal. App.3d 584 [ 134 Cal. Rptr. 39 , 91 A.L.R.3d 759 ]. In that case the defendant was convicted of violating Los Angeles Municipal Code section 43.30, which prohibited advertising or engaging in the telling of fortunes and related activities. The defendant insisted that she was a priestess of Wicca, the religion of witchcraft, and that the Los Angeles ordinance unconstitutionally interfered with the practice of her religion and restricted her freedom of speech. The Court of Appeal cited Los Angeles Municipal Code section 43.31, which exempts from section 43.30 any legitimate religious practices, and emphasized that the jury had been instructed that in order to convict it must find the defendant's fortunetelling to be a business, not a religious practice. Since the jury found a violation, the appellate court deemed the defendant's activity to be a business, subject to regulation: "The constitutional right to freedom of speech does not prevent the Legislature from regulating or prohibiting commercial enterprises which are harmful to the public welfare." ( Id., at p. 591.)

We are unable to subscribe to Bartha's broad characterization of fortunetelling as an exclusively commercial activity, and to the theory that it therefore can be indiscriminately regulated, or, in this instance, wholly prohibited. (1) The essence of the issue whether an activity falls within the constitutional protection of "speech" is whether the "speaker," by engaging in the activity, is communicating information of any sort. For example, in Powers v. Floersheim (1967) 256 Cal. App.2d 223 [ 63 Cal. Rptr. 913 ], the defendant urged that printed forms used in collecting debts were constitutionally protected, and thus their distribution could not be regulated. The Court of Appeal determined that the forms did not constitute a type of speech: "No opinion, thought expression, or other form of information is contained in the forms under discussion. They are merely tools of a trade, much as a hammer is a tool of the trade of carpentry. The purchaser seeks no information from the form, and the designer seeks to convey none." ( Id. at p. 233.)

Fortunetelling is different. It involves the communication of a message directly from the fortuneteller to the recipient. That words are used is not critical; the key is that the words convey thoughts, opinions and, sometimes, fiction and falsehoods. [3] This communication between persons, however, is at the very core of what is known as speech. That fortunetelling consists of speech does not of itself determine what level of protection it must be afforded under article I, section 2, of the Constitution, but it does establish that fortunetelling is not a "mere commercial activity."

The conclusion in Bartha implies that to characterize an activity as merely commercial magically removes any constitutional barriers to its regulation or prohibition. (2) But it is manifest that speech does not lose its protected character when it is engaged in for profit. ( Va. Pharmacy Bd. v. Va. Consumer Council (1976) 425 U.S. 748, 761 [ 48 L.Ed.2d 346, 358 , 96 S.Ct. 1817 ]; Time, Inc. v. Hill (1967) 385 U.S. 374, 397 [ 17 L.Ed.2d 456, 472 , 87 S.Ct. 534 ]; New York Times Co. v. Sullivan (1964) 376 U.S. 254, 266 [ 11 L.Ed.2d 686, 698 , 84 S.Ct. 710 , 95 A.L.R.2d 1412 ]; Joseph Burstyn, Inc. v. Wilson (1952) 343 U.S. 495, 501-502 [ 96 L.Ed. 1098, 1106 , 72 S.Ct. 777 ]; People v. Glaze (1980) 27 Cal.3d 841, 846 [ 166 Cal. Rptr. 859 , 614 P.2d 291 ].) "It should be remembered that the pamphlets of Thomas Paine were not distributed free of charge.... Freedom of speech, freedom of the press, freedom of religion are available to all, not merely to those who can pay their own way." ( Murdock v. Pennsylvania (1943) 319 U.S. 105, 111 [ 87 L.Ed. 1292, 1297 , 63 S.Ct. 870 , 146 A.L.R. 81 ].) Further, "Characterizing [a] publication as a business, and the business as a nuisance, does not permit an invasion of the constitutional immunity against restraint." ( Near v. Minnesota (1931) 283 U.S. 697, 720 [ 75 L.Ed. 1357, 1370 , 51 S.Ct. 625 ].) "The idea is not sound ... that the First Amendment's safeguards are wholly inapplicable to business or economic activity. And it does not resolve where the line shall be drawn in a particular case merely to urge ... that an organization for which the rights of free speech and free assembly are claimed is one `engaged in business activities' or that the individual who leads it in exercising these rights receives compensation for doing so." ( Thomas v. Collins (1944) 323 U.S. 516, 531 [ 89 L.Ed. 430, 441 , 65 S.Ct. 315 ].)

It is thus no answer to the charge that the ordinance violates the Constitution to characterize what it prohibits as commercial activity; we must still inquire whether speech is being repressed. And when, as here, speech is indeed involved, we must evaluate the constitutionality of the regulation under the stringent tests of article I, section 2, of our Constitution.

The City argues that even if fortunetelling can be characterized as speech, it is commercial speech which is not entitled to the same high level of protection as noncommercial speech. In order to determine the proper test by which to evaluate the ordinance under the Constitution, we must thus determine whether fortunetelling for consideration involves commercial or noncommercial speech.

There is no clearly articulated test to determine what constitutes commercial speech. Rather, phrases from certain opinions of the United States Supreme Court have been used to evaluate types of speech. Thus commercial speech has been referred to as "speech which does `no more than propose a commercial transaction'" ( Va. Pharmacy Bd. v. Va. Consumer Council, supra, 425 U.S. 748, 762 [ 48 L.Ed.2d 346, 358 ], quoting from Pittsburgh Press Co. v. Human Relations Comm'n (1973) 413 U.S. 376, 385 [ 37 L.Ed.2d 669, 677 , 93 S.Ct. 2553 ]), and as "expression related solely to the economic interests of the speaker and its audience." ( Central Hudson Gas & Ele. Corp. v. Public Serv. Com'n (1980) 447 U.S. 557, 561 [ 65 L.Ed.2d 341, 348 , 100 S.Ct. 2343 ].) The City concedes that the cases dealing with commercial speech all involve solicitation of sales and advertising that is, speech proposing a commercial transaction. However, it urges that although fortunetelling for profit does not propose a transaction, it provides the mechanism for completing the transaction, and thus relates only to the economic interests of the parties.

This theory of commercial speech has no basis in precedent and fundamentally misconstrues the commercial-noncommercial distinction. Commercial speech cases have involved such activity as in-person solicitation by attorneys to obtain remunerative employment ( Ohralik v. Ohio State Bar Assn. (1978) 436 U.S. 447 [ 56 L.Ed.2d 444 , 98 S.Ct. 1912 ]), "the mere solicitation of patronage implicit in a trade name" ( Friedman v. Rogers (1979) 440 U.S. 1, 11-12, fn. 10 [ 59 L.Ed.2d 100, 111 , 99 S.Ct. 887 ]), and negligently untruthful advertising ( People v. Superior Court (Olson) (1979) 96 Cal. App.3d 181 [ 157 Cal. Rptr. 628 ], cert. den. (1980) 446 U.S. 935 [ 64 L.Ed.2d 787 , 100 S.Ct. 2152 ]). In contrast, cases held not to involve merely commercial speech have dealt with advertisements that contain factual material going beyond the proposition of a commercial transaction ( Bigelow v. Virginia (1975) 421 U.S. 809 [ 44 L.Ed.2d 600 , 95 S.Ct. 2222 ]), solicitation by attorneys for representation without charge in order to advance political beliefs ( In re Primus (1978) 436 U.S. 412 [ 56 L.Ed.2d 417 , 98 S.Ct. 1893 ]), and the sale on the streets of maps to movie stars' homes ( Welton v. City of Los Angeles (1976) 18 Cal.3d 497 [ 134 Cal. Rptr. 668 , 556 P.2d 1119 ]).

(3) The principle emerging from these cases is that commercial speech is that which has but one purpose to advance an economic transaction. By contrast, noncommercial speech encompasses activities extending beyond that purpose. For example, an advertisement that cherries can be purchased for a dollar a box at store X may be commercial speech, but an advertisement informing the public that the cherries for sale at store X were picked by union workers is more: it communicates a message beyond that related to the bare economic interests of the parties.

The act of telling fortunes goes beyond the mere proposal of a transaction. It involves the passing of ideas and information some valid, some questionable, some false between the fortuneteller and the client. This exchange is unrelated to any consideration the client pays to receive the communication the consideration is not the object of the communication. When Stevens charges a fee for fortunetelling, she sells her services, not an advertisement of her services. (See Welton v. City of Los Angeles, supra, 18 Cal.3d 497, 503 [plaintiff's street-vending of maps did not constitute advertising: she was selling printed material, not its advertisement].) If we were to accept the City's theory, a lecture for or against Marxism, abortion, nuclear power, or racial supremacy would be commercial speech if people paid an admission charge to hear it, because the lecture would complete the transaction. Such a result would be unprecedented and untenable.

(4) The constitutional right of free expression is an essential ingredient of our democratic society. "It is designed and intended to remove governmental restraints from the arena of public discussion, putting the decision as to what views shall be voiced largely into the hands of each of us, in the hope that use of such freedom will ultimately produce a more capable citizenry and more perfect polity and in the belief that no other approach would comport with the premise of individual dignity and choice upon which our political system rests." ( Cohen v. California (1971) 403 U.S. 15, 24 [ 29 L.Ed.2d 284, 293 , 91 S.Ct. 1780 ]; accord, Thornhill v. Alabama (1940) 310 U.S. 88, 97 [ 84 L.Ed. 1093, 1099 , 60 S.Ct. 736 ].) The airing of opposing views is fundamental to an informed electorate and, through it, a free society.

(5) The City maintains that fortunetelling does not fall within the protection of the Constitution because it does not concern or affect the political process. But this argument fails to comprehend the broad scope and purpose of the constitutional guarantee of free speech. We explained the reach of this principle in In re Giannini (1968) 69 Cal.2d 563, 569-570 , footnote 3 [ 72 Cal. Rptr. 655 , 446 P.2d 535 ], certiorari denied (1969) 395 U.S. 910 [ 23 L.Ed.2d 223 , 89 S.Ct. 1743 ], overruled on other grounds in Crownover v. Musick (1973) 9 Cal.3d 405, 431 [ 107 Cal. Rptr. 681 , 509 P.2d 497 ]: "Unquestionably, [there is] First Amendment protection for communication that in fact exhibits no special relationship to the political process. Even to the extent that it is possible, for example, to isolate and enumerate precise opinions or concepts in a Joyce or Shakespeare, the ideas they espouse might well have little apparent relevance to the political process. Yet no one could doubt that these works warrant First Amendment protection. [¶] One rationale for this result is that, although the First Amendment is designed to protect only communication that forms the basis for workable democracy in the exchange of ideas relevant to political decisions, `[t]he line between the informing and the entertaining is too elusive' [citation] and courts must therefore cast a wide net over all forms of communication in order to protect that which is of potential political relevance. An equally persuasive rationale, however, is that the life of the imagination and intellect is of comparable import to the presentation of the political process; the First Amendment reaches beyond protection of citizen participation in, and ultimate control over, governmental affairs and protects in addition the interest in free interchange of ideas and impressions for their own sake, for whatever benefit the individual may gain." (See also People v. Glaze, supra, 27 Cal.3d 841, 849 .)

Under either rationale, fortunetelling deserves protection. First, it is impossible to say that fortunetellers impart no political message in their communications. In their vision of the future there may be a view of society as they perceive it may one day be. Such a communication conceivably could contain the spark of a political flame. Second, fortunetelling may fire the imagination and stimulate discussion of the future. That some even a majority may find this mode of communication distasteful, ridiculous or even corrupt is irrelevant to constitutional concerns. ( Murdock v. Pennsylvania, supra, 319 U.S. 105, 115-116 [ 87 L.Ed. 1292, 1300 ]; Morris v. Municipal Court (1982) 32 Cal.3d 553, 566-567 [ 186 Cal. Rptr. 494 , 652 P.2d 51 ]; Welton v. City of Los Angeles, supra, 18 Cal.3d 497, 504 .) Thus any prohibition or regulation of fortunetelling must be given full scrutiny under the Constitution.

(6) However, it is clear that even though a communication is potentially entitled to constitutional protection, it is not necessarily immune from regulation. ( Konigsberg v. State Bar (1961) 366 U.S. 36, 49 [ 6 L.Ed.2d 105 , 116, 81 S.Ct. 997 ].) Some forms of speech have been held to fall completely outside the protected arena. ( Id. at p. 50 [ 6 L.Ed.2d at p. 116 ].) In addition, a regulation of speech may be upheld if it passes certain stringent tests ( id. at pp. 50-51 [6 L.Ed.2d at pp. 116-117]) that we outline below.

One of the most difficult questions in a free speech analysis is which test must be applied to the particular regulation and communication before the court. A helpful explanation of the United States Supreme Court's view of the matter has been suggested by one scholar: "The Supreme Court has evolved two distinct approaches to the resolution of first amendment claims; the two correspond to the two ways in which government may `abridge' speech. If a government regulation is aimed at the communicative impact of an act, ... [the] regulation is unconstitutional unless government shows that the message being suppressed poses a `clear and present danger,' constitutes a defamatory falsehood, or otherwise falls on the unprotected side of one of the lines the Court has drawn to distinguish those expressive acts privileged by the first amendment from those open to government regulation with only minimal due process scrutiny. If a government regulation is aimed at the noncommunicative impact of an act ... [the] regulation is constitutional, even as applied to expressive conduct, so long as it does not unduly constrict the flow of information and ideas. [Here] ... the `balance' between the values of freedom of expression and the government's regulatory interests is struck on a case-by-case basis, guided by whatever unifying principles may be found in past decisions." (Tribe, American Constitutional Law (1978) p. 582.)

It is unnecessary, however, for us to determine whether the ordinance is "aimed at the communicative element" of the prohibited speech, for it is unconstitutional under either test.

(7) There are four categories of speech that are held not entitled to First Amendment protection. These are fighting words ( Chaplinsky v. New Hampshire (1942) 315 U.S. 568 [ 86 L.Ed. 1031, 1034 , 62 S.Ct. 766 ], obscenity ( Roth v. United States (1957) 354 U.S. 476 [ 1 L.Ed.2d 1498 , 77 S.Ct. 1304 ]), defamatory falsehoods ( Gertz v. Robert Welch, Inc. (1974) 418 U.S. 323 [ 41 L.Ed.2d 789 , 94 S.Ct. 2997 ]), and speech carrying a clear and present danger of incitement to violence ( Brandenburg v. Ohio (1969) 395 U.S. 444 [ 23 L.Ed.2d 430 , 89 S.Ct. 1827 ]). "It has been well observed that such utterances are no essential part of any exposition of ideas, and are of such slight social value as a step to truth that any benefit that may be derived from them is clearly outweighed by the social interest in order and morality." ( Chaplinsky v. New Hampshire, supra, 315 U.S. 568, 572 [ 86 L.Ed. 1031, 1035 ].)

The City contends that fortunetelling falls in a category of speech that is not entitled to free speech protection: it is within the legislative body's power to determine that fortunetelling is inherently deceptive, and when as here the legislative body has done so, the speech is not protected. The City cites Gertz v. Robert Welch, Inc., supra, 418 U.S. 323, 339-340 [ 41 L.Ed.2d 789, 805 ]: "Under the First Amendment there is no such thing as a false idea. However pernicious an opinion may seem, we depend for its correction not on the conscience of judges and juries but on the competition of other ideas. But there is no constitutional value in false statements of fact." The City further directs our attention to In re Bartha, supra, 63 Cal. App.3d 584 , 591: "The ordinance need not necessarily be limited to cases involving an actual intent to defraud. It is within the police power of the municipality and province of the legislative body to determine that the business of fortune-telling is inherently deceptive and that its regulation or prohibition is required in order to protect the gullible, superstitious, and unwary." [4]

(8) First it must be emphasized that the ordinary deference a court owes to any legislative action vanishes when constitutionally protected rights are threatened. "The rational connection between the remedy provided and the evil to be curbed, which in other contexts might support legislation against attack on due process grounds, will not suffice." ( Thomas v. Collins, supra, 323 U.S. 516, 530, 531-532 [ 89 L.Ed. 430, 440-441 , 65 S.Ct. 315 ]; see also N.A.A.C.P. v. Button (1963) 371 U.S. 415, 438-439 [ 9 L.Ed.2d 405, 421 , 83 S.Ct. 328 ]; Wirta v. Alameda-Contra Costa Transit Dist. (1967) 68 Cal.2d 51, 60 [ 64 Cal. Rptr. 430 , 434 P.2d 982 ].) Thus we would abandon our constitutional duty if we took at face value the municipality's determination that fortunetelling is inherently deceptive and not protected by the state constitutional free speech protection.

Fraudulent deceit, actionable as a tort, is defined in part by California Civil Code section 1710 as "1. The suggestion, as a fact, of that which is not true, by one who does not believe it to be true; [¶] 2. The assertion, as a fact, of that which is not true, by one who has no reasonable ground for believing it to be true." Predictions of the future have been characterized as expressions of opinion, not actionable unless the speaker knows the opinion is unwarranted or induces reliance on the opinion as if it were an expression of fact. ( Richard P. v. Vista Del Mar Child Care Service (1980) 106 Cal. App.3d 860, 865-866 [ 165 Cal. Rptr. 370 ].)

It must be conceded that many persons practicing the "art" of fortunetelling are engaging in fraudulent activity. Many fortunetellers have no belief in their powers to predict the future. If such persons obtain consideration for their services, the activity could be deemed fraudulent; their statements would be false statements of fact, for they would not be expressing what they truly believe will happen in the future. However, it is also true that some persons believe they possess the power to predict what has not yet come to pass. When such persons impart their beliefs to others, they are not acting fraudulently; they are communicating opinions which, however dubious, are unquestionably protected by the Constitution.

It must also be noted that there are many persons other than professional fortunetellers who purport to predict the future: e.g., astrology columnists in daily newspapers, economists who prognosticate interest rates and other business conditions, investment counsellors who forecast stock market trends, sportswriters and oddsmakers who predict the winners of athletic contests, horserace handicappers, pollsters who forecast election returns, and clergymen who describe the concept of a hereafter.

The City maintains it is within the legislative province to prohibit fraudulent fortunetelling. This argument, however, does not save the ordinance. (9) It is true that a state may protect its citizens from fraud. ( Cantwell v. Connecticut (1940) 310 U.S. 296, 306 [ 84 L.Ed. 1213, 1219 , 60 S.Ct. 900 , 128 A.L.R. 1352 ]; Schneider v. State (1939) 308 U.S. 147, 164 [ 84 L.Ed. 155, 166 , 60 S.Ct. 146 ]; Marks v. City of Roseburg (1983) 65 Ore.App. 102 [ 670 P.2d 201, 204 ].) Yet "Broad prophylactic rules in the area of free expression are suspect. [Citations.] Precision of regulation must be the touchstone in an area so closely touching our most precious freedoms." ( N.A.A.C.P. v. Button, supra, 371 U.S. 415, 438 [ 9 L.Ed.2d 405, 421 ]; In re Primus, supra, 436 U.S. 412, 434 [ 56 L.Ed.2d 417, 436 , 98 S.Ct. 1893 ]; Erznoznik v. City of Jacksonville (1975) 422 U.S. 205, 217-218 [ 45 L.Ed.2d 125, 135-136 , 95 S.Ct. 2268 ]; Morris v. Municipal Court, supra, 32 Cal.3d 553, 565 .) Regulation, "whether aimed at fraud or other abuses, must not trespass upon the domains set apart for free speech and free assembly." ( Thomas v. Collins, supra, 323 U.S. 516, 532 [ 89 L.Ed. 430, 441 ]; Young v. Municipal Court (1971) 16 Cal. App.3d 766, 769 [ 94 Cal. Rptr. 331 ].) A regulation is suspect, therefore, if it prohibits protected expression, even though it also guards the public from fraud.

The justification for this rule is well recognized: when a statute or ordinance is overbroad, condemning protected as well as unprotected speech, communication of the protected speech is effectively chilled. It is not enough to say that the enactment is aimed at unprotected speech; there remains the real danger that a statute that also encompasses protected speech will be used to punish or prohibit its exercise. (See Thornhill v. Alabama, supra, 310 U.S. 88, 97-98 [ 84 L.Ed. 1093, 1099 ].) And it is irrelevant that much of the activity prohibited by the statute is unprotected. ( Near v. Minnesota, supra, 283 U.S. 707 , 720 [ 75 L.Ed. 1357, 1369 , 51 S.Ct. 625 ].)

(10) The City assures us that the ordinance is aimed only at communications that purport to predict future events. Assuming that such a ban would be permissible, however, the ordinance contains no words to this effect. Thus the prohibition against "spiritual reading" could encompass Bible lessons, the bar against "hypnotism" could include hypnosis as an accepted technique of the psychotherapist, the banning of "magic" could prevent numerous popular theatrical performances, and the prohibition of "prophecy" could interfere with many religious services. The ordinance is clearly overbroad, applying to many activities that are protected by the California Constitution.

Thus the ordinance cannot be saved as a valid regulation of unprotected speech. It remains to be determined whether the ordinance is a valid regulation of something other than the communicative element of speech, with only an incidental effect on that which is protected by the Constitution.

At least on its face, the ordinance does not purport to be aimed at the communicative element of the speech it bans. That is, the City claims it does not prohibit the message that fortunetellers wish to convey; it seeks to prohibit only the fraud of telling fortunes with the sole intent of bilking the gullible by trading compensation for falsehoods. (11a) But we need not resolve this question, for even if the ordinance is not aimed at the communicative element, it is unconstitutional.

(12) In United States v. O'Brien (1968) 391 U.S. 367 [ 20 L.Ed.2d 672 , 88 S.Ct. 1673 ], the Supreme Court set out a four-part test to determine the constitutionality of governmental enactments that regulate nonspeech elements of conduct and have an incidental effect on protected communications: "a government regulation is sufficiently justified if it is within the constitutional power of the Government; if it furthers an important or substantial governmental interest; if the governmental interest is unrelated to the suppression of free expression; and if the incidental restriction on alleged First Amendment freedoms is no greater than is essential to the furtherance of that interest." ( Id., at p. 377 [ 20 L.Ed.2d at p. 680 ].)

The first two elements of this test provide no difficulty for the ordinance. The government may constitutionally enact laws protecting the public welfare ( Schneider v. State, supra, 308 U.S. 147, 160 [ 84 L.Ed. 155, 164 , 60 S.Ct. 146 ]), and the protection of the citizens from fraud is an important state interest. (See ante at p. 515.) The third element of the test, it has been said, determines what approach a court will take to the evaluation of the regulation's constitutionality, for here it must be asked whether the regulation focuses on the communicative element of the speech it prohibits. If the regulation does so focus, it will be invalidated unless the speech is unprotected. If the regulation does not so focus, the fourth step of the test must be applied. (Ely, Flag Desecration: A Case Study in the Roles of Categorization and Balancing in First Amendment Analysis (1975) 88 Harv.L.Rev. 1482, 1484, 1496-1497.)

(11b) Having discussed the invalidity of the ordinance under the former approach, we now turn to an examination under the latter. This step of the O'Brien analysis involves a balancing of interests and a determination that government has used the "least restrictive means" to regulate the conduct in question. ( Morris v. Municipal Court, supra, 32 Cal.3d 553, 565 .)

First, we must "weigh the circumstances and ... appraise the substantiality of the reasons advanced in support of the regulation of the free enjoyment of the rights." ( Schneider v. State, supra, 308 U.S. 147, 161 [ 84 L.Ed. 155, 165 ]; Dillon v. Municipal Court (1971) 4 Cal.3d 860, 869 [ 94 Cal. Rptr. 777 , 484 P.2d 945 ].) The City argues that the public's substantial interest in preventing fraud outweighs the interest of an individual who wishes to express the type of communication that the ordinance bans. The ordinance constitutes only a minimal imposition on the free speech rights of fortunetellers, contends the City, for it prohibits fortunetelling only when practiced for compensation.

As we have seen, however, the fact that speech is engaged in for profit does not cause it to lose its constitutional protection. (Part I, ante. ) What is more, a regulation banning certain speech only when expressed for profit cannot be dismissed as imposing but a minor restriction on that speech. In Murdock v. Pennsylvania, supra, 319 U.S. 105 , the United States Supreme Court invalidated an ordinance imposing a license fee on door-to-door solicitors, including persons distributing religious information: "Those who can tax the privilege of engaging in this form of missionary evangelism can close its doors to all those who do not have a full purse." ( Id., at p. 112 [ 87 L.Ed. at p. 1298 ].) In Schaumburg v. Citizens for a Better Environ. (1980) 444 U.S. 620 [ 63 L.Ed.2d 73 , 100 S.Ct. 826 ], the court struck down an ordinance denying charitable organizations a permit to solicit door-to-door unless the organization could guarantee that 75 percent of the funds collected would be allocated to the charitable purpose. In Minneapolis Star v. Minnesota Comm'r of Rev. (1983) 460 U.S. 575 [ 75 L.Ed.2d 295 , 103 S.Ct. 1365 ], the court held invalid a tax that affected only newspapers. And in Weaver v. Jordan (1966) 64 Cal.2d 235 [ 49 L.Ed.2d 537 , 411 P.2d 289 ], certiorari denied (1966) 385 U.S. 844 [ 17 L.Ed.2d 75 , 87 S.Ct. 49 ], we held that an enactment banning pay television was unconstitutional: "The assertion of defendant and of amici curiae that the Act does not invade freedom of expression because it does not prohibit subscription television, but merely forbids direct charges for programs transmitted to the home, is devoid of substance. The trial court correctly observed that `This is comparable to asserting that no prohibition of expression would exist in the case of newspapers or motion pictures if a statute were adopted requiring their distribution or showing without charge.'" ( Id., at p. 246.) Although we cannot agree with Stevens and the Church that the ordinance constitutes a total ban on their speech, the substantial degree to which it restricts that speech cannot be ignored.

Against the degree to which the ordinance affects speech we must balance the City's interest in regulating fraud the asserted justification of the ordinance. But it is not simply the City's interest in banning fraud that goes into the balance. It is a balance "at the margin that is, [a court] must balance no more than the state's interest in the added effectiveness of the chosen means against the individual interest in the use of less drastic ones." (Italics added.) (Comment, Less Drastic Means and the First Amendment (1969) 78 Yale L.J. 464 , 467-468.) To determine this interest we must inquire whether less drastic means exist to prohibit such fraud.

A law prohibiting fraud in fortunetelling could be written; indeed, it exists. Penal Code section 332 provides that "Every person who by ... pretensions to fortunetelling, trick, or other means whatever ... fraudulently obtains from another person money or property of any description, shall be punished as in case of larceny of property of like value." Such a law prohibits unprotected fraudulent fortunetelling while allowing true believers to practice their art. The public goal is vindicated without unduly suppressing the exercise of constitutional rights.

It is irrelevant that punishing fraud may be less convenient than prohibiting all situations in which a potential for fraud arises. ( Schneider v. State, supra, 308 U.S. 147, 164 [ 84 L.Ed. 155, 166 ].) Similarly, "Mere legislative preference for one rather than another means for combatting substantive evils" is not sufficient to justify an overbroad ordinance. ( Thornhill v. Alabama, supra, 310 U.S. 88, 95-96 [ 84 L.Ed. 1093, 1098 ].) False and fraudulent statements are inevitable in a society in which freedom of speech is paramount. ( New York Times Co. v. Sullivan, supra, 376 U.S. 254, 271-272 [ 11 L.Ed.2d 686, 701 ].) "But the people of this nation have ordained in the light of history, that, in spite of the probability of excesses and abuses, these liberties are, in the long view, essential to enlightened opinion and right conduct on the part of the citizens of a democracy." ( Cantwell v. Connecticut, supra, 310 U.S. 296, 310 [ 84 L.Ed. 1213, 1221 ].)

Thus there are methods to prevent fraudulent fortunetelling that would impose a less drastic restriction on protected speech. Any marginal interest the City may have in more complete and convenient regulation in this area by a total ban on fortunetelling for compensation is outweighed by the interest in free and open speech on all subjects. Therefore the ordinance fails both the least drastic means test and the balancing of interests tests. We must hold it unconstitutional under article I, section 2, of the California Constitution.

We rely on article I, section 2, of the California Constitution, which declares in part that "Every person may freely speak, write and publish his or her sentiments on all subjects, being responsible for the abuse of this right. A law may not restrain or abridge liberty of speech or press." (13) This provision is "more definitive and inclusive than the First Amendment." ( Wilson v. Superior Court (1975) 13 Cal.3d 652, 658 [ 119 Cal. Rptr. 468 , 532 P.2d 116 ].)

The question remains whether the ordinance can be "saved" by judicial construction. (14) In Welton v. City of Los Angeles, supra, 18 Cal.3d 497 , we reviewed the rules governing the construction of an overbroad statute to avoid conflict with the Constitution: "First, the enactment may be validated if its terms are reasonably susceptible to an interpretation consistent with the Constitution. [Citations.] Second, the court should construe the enactment so as to limit its effect and operation to matters that may be constitutionally regulated or prohibited. [Citations.] [¶] Both principles are limited by the further rule that judicial construction must not create uncertainty inhibiting exercise of a constitutional right." ( Id., at pp. 505-506.) The ordinance at issue in Welton was aimed principally at the sale of "goods, wares and merchandise" on the streets, and affected only incidentally the sale of printed matter. Thus, the reference to printed matter could be severed in order to save the constitutionality of the ordinance without substantially altering its focus. ( Id., at pp. 506-507.)

In contrast to Welton is Dillon v. Municipal Court, supra, 4 Cal.3d 860 . There the City of Seaside delegated the duty of issuing parade permits to the Chamber of Commerce. We held the ordinance governing permits unconstitutional because it gave the granting official unlimited discretion. ( Id., at pp. 869-870.) Further, we determined that we had "no right to add to this ordinance provisions which it obviously lacks." ( Id., at p. 871.)

(15) Although the ordinance in the case at bar contains an obvious defect that could be cured by limiting its reach to fraudulent activity carried out by fortunetellers, we would abuse our power by adding such language to the measure. This is not a case like Welton , in which the ordinance could stand on its own without the invalid reference. Before us now is an enactment that is invalid in its essence; we would be required to rewrite it in order to save it. This we may not do.

For the reasons stated we hold the ordinance invalid under article I, section 2, of the California Constitution.

The order is reversed.

Bird, C.J., Broussard, J., and Grodin, J., concurred.

KAUS, J.

I agree that the ordinance is overbroad. On the other hand, the First Amendment has been around for a long time, as have been state and local laws against fortunetelling and decisions upholding them. (See cases cited in In re Bartha (1976) 63 Cal. App.3d 584, 589 [ 134 Cal. Rptr. 39 , 91 A.L.R.3d 759 ].) It is a fact that plaintiff cites no case which elevates fortunetelling to the same free speech pedestal as does the majority. I cannot help feeling that the core values of the First Amendment have somehow become obliterated in the court's somewhat formalistic application of precedent based on entirely different facts. In short, I am more optimistic than the majority that a constitutionally acceptable ordinance can be drafted.

Reynoso, J., concurred.

LUCAS, J., Concurring and Dissenting.

I concur in the judgment on the limited ground that Azusa's fortunetelling ordinance is unduly broad. By its terms, the ordinance could apply to (and flatly prohibit) such bona fide practices as therapeutic hypnotism by a trained professional, or religious prophecy by a minister or preacher paid by his congregation to give sermons. In its commendable zeal to list every possible kind of practitioner of the various "occult arts," the city unfortunately included some persons whose practices are not likely to result in defrauding the public.

I dissent, however, to the majority's alternative holding that First Amendment principles would preclude the city from prohibiting such inherently deceptive and fraudulent practices as, for example, tea reading, crystalgazing, necromancy or fortunetelling for money. We may take judicial notice of the fact that such devices are routinely, if not uniformly used to bilk or fleece gullible patrons. As the majority concedes, "many persons practicing the `art' of fortunetelling are engaging in fraudulent practices," having no belief in their power to predict the future. ( Ante, p. 515.) The majority suggests, however, that "some persons believe they possess the power to predict" the future ( ibid. ), and that the First Amendment protects their "opinions," even when they exact a fee therefor.

With due respect, the majority is far too naive in its assumption. The majority's unsubstantiated concept of the guileless seer, accepting money in exchange for bona fide attempts at prognostication, stands in direct conflict with the traditional, and much more realistic, appraisal that "the business of fortunetelling is inherently deceptive," and that "its regulation or prohibition is required in order to protect the gullible, superstitious, and unwary. [Citations.]" ( In re Bartha (1976) 63 Cal. App.3d 584, 591 [ 134 Cal. Rptr. 39 , 91 A.L.R. 3d 759 ].)

Like most Southern California cities, Azusa undoubtedly attracts a fair share of elderly, retired citizens who, by reason of their advanced age or infirmity, are easy prey for those who solicit a fee for such "services" as predicting the future, communicating with deceased loved ones, and so on. These citizens, typically living on limited or fixed incomes, are unfortunately often the group most easily duped yet least able to afford the consequences thereof. Just as a community can protect its citizens from their own cupidity by passing antigambling ordinances, Azusa may protect its citizens from their own gullibility by passing an antifortunetelling ordinance.

The majority's suggestion that existing criminal fraud statutes are an adequate remedy seems patently incorrect. First, such statutes probably would not afford a remedy in the absence of proof of an intent to deceive, a matter often quite difficult to establish beyond a reasonable doubt. Second, unlike a prohibition upon the practice itself, the penal laws operate only after it has occurred and the fraud discovered. Many victims of "occult" swindlers are reluctant (through embarrassment or naivety) to complain to the authorities regarding the fraud. In addition, the availability of a possible criminal sanction would be of little solace to an Azusa widow bilked of her life savings by a necromancer or crystalgazer who had the "foresight" to take early leave.

Balanced against the compelling interest in protecting gullible citizens from fraud, the fortuneteller's supposedly "protected" right to charge a fee for giving an "opinion" as to future events pales to insignificance. I would uphold a narrowly drawn prohibitory ordinance as against a constitutional challenge based on free speech principles.

[1] Throughout this opinion we use the term "fortunetelling" to describe the communication prohibited by the ordinance, because plaintiffs challenge the ordinance insofar as it affects the telling of fortunes. Our analysis, however, applies to all the numerous related and unrelated activities banned by the ordinance.

When the term "First Amendment" is used, it is merely illustrative and employed as a shorthand reference to article I, section 2, of the California Constitution.

[2] We assume that plaintiffs meant to ask the court to enjoin the City from preventing them from conducting such services.

[3] For example, a person who uses words to sell a worthless piece of land may be prosecuted without violating his free speech rights: such words are used only to conduct the transaction and carry no value in and of themselves. Fortunetelling, on the other hand, involves the expression of ideas, whether through words, actions or signs. As such, it is protected speech.

[4] The City also cites cases involving the regulation of deceptive commercial speech. (E.g., Central Hudson Gas and Elec. v. Public Serv. Comm., supra, 447 U.S. 557 ; Ohralik v. Ohio State Bar Assn., supra, 436 U.S. 447 .) These cases are inapposite, however, as fortunetelling is not commercial speech per se. (See part II, ante. )