HECHT, Justice, joined by COOK and CORNYN, Justices, concurring in the judgment.

I join in the Court's judgment, but not in its opinion. I agree that mandamus should issue directing the respondent district court to vacate the gag orders of which relator complains, but not to reinstate relator as guardian ad litem. I also agree that we should not direct the district court to allow relator access to court records when relator has failed to establish that her access to those records has been restricted. I differ with the Court's reasons for these decisions, however, and write to explain why.

I

Deciding whether the two gag orders which the district court issued in the pending litigation are invalid is not the principal occupation of the Court's opinion. Those orders have already been ordered vacated in response to relator's request for emergency relief. 837 S.W.2d 73. Our reasons for granting this relief do not require elaborate explanation. The Court is unanimous in the view that the gag orders are invalid because they are too broad, they are not necessary to protect against an imminent threat to the administration of justice, and they were issued without following procedures to safeguard against suppression of relator's constitutionally protected speech. Issuance of the orders was a clear abuse of discretion from which relator has no adequate remedy by appeal. Thus, relator is entitled to mandamus relief to have the orders set aside.

That is the Court's decision. Most of the Court's opinion is spent defending its efforts to decide relator's free speech claims using only article I, section 8 of the Texas Constitution without recourse to the First Amendment to the United States Constitution. Borrowing from the literature of the "new federalism movement" of the past fifteen years, the Court adopts a method of constitutional analysis by which it examines the Texas Constitution first, and if a right is found to be protected, never reaches the federal constitution question. In theory, the Court's methodology contemplates that federal law construing a federal constitutional provision will be instructive but not controlling in construing a corresponding provision of the state constitution; in actuality, the Court attempts to ignore federal law altogether. Accordingly, it ventures an independent examination and application of article I, section 8, with no argument and little briefing by the parties, and without regard to more fully developed First Amendment law. Then after determining that the language of article I, section 8 is different and highly distinctive, the Court shortly concludes that the test for reviewing gag orders under that provision is one which happens to be identical to the test under the First Amendment. The Court follows this analysis with a lengthy apologia and accolade for its new method.

The Court's approach to this case, it seems to me, is contrived and unnecessarily extreme. The Court goes to great lengths to decide this case on our state constitution alone, even though the result would be the same under the First Amendment, for the same reasons. Although a state constitutional provision should not be ignored simply because it has a federal analogue, I think the converse is equally true: federal constitutional law should not be ignored simply because there exists a related state constitutional provision. Where, as here, the issue raised can be resolved on First Amendment grounds entirely consistent with the Texas Constitution, there is every reason to do so. What reasons the Court may have for avoiding this straightforward course is a question to which I shall return.

A

The pending litigation involves claims by over 200 children and their parents for injuries due to exposure to toxic chemicals. The parents settled their claims five years ago, and defendants have proposed to settle the children's claims, subject to the district court's approval. Relator, the guardian ad litem for the children, opposes the settlement. At a hearing on the proposed settlement on August 23, 1991, the district court apparently became concerned that differences and misunderstandings among the participants in the litigation threatened the proposed settlement and the best interests of the minor plaintiffs. Without a request from any party, the district court issued the first of two gag orders, orally instructing plaintiffs and defendants, their counsel, and relator not to discuss the case outside the courtroom. The court also "abated" relator's appointment as guardian ad litem and directed her to have no contact with plaintiffs pending further order of the court. Several days later, on September 11, the district court committed its order to writing, explaining the reasons for its issuance. This second, written, order stated in substance:

BE IT REMEMBERED THAT ON THE 23RD DAY OF AUGUST, 1991, at a hearing in these consolidated cases, counsel for Plaintiffs, counsel for Defendant, and the then Guardian ad Litem were present.

Upon hearing evidence that conflicts between counsel and the parents of the minor children were resulting in miscommunications with the parents of the children and with the media and general public, this Court, on its own motion, issued a protective order in the best interest of the minor children of this suit. In so doing, the Court found there was a need for such an extraordinary remedy and ORDERED counsel as follows:

Counsel in this case, present and former, are expressly ORDERED to refrain from discussing or publishing in writing or otherwise, any matters of this case with any persons other than their clients, agents, or employees in the necessary course of business in this case.

Counsel is ORDERED to refrain from any public comment, casual or otherwise concerning the facts of this case or the conduct of counsel in this case other than in a court hearing.

Counsel is ORDERED to inform their clients, witnesses, agents and representatives that this ORDER extends to each of them and is subject to a finding of contempt by this court from disobedience, direct or indirect comment intended to violate this ORDER. Counsel was and is directed to communicate with their clients only, and advise each that they are directed to refrain from discussing the case except with counsel.

At that hearing, the Court abated the appointment of the Guardian ad Litem until further Order of this Court, but specifically applied this protective order to the Guardian and that during the abatement, the Guardian was to have no contact with Plaintiffs. On September 11, 1991, the Court dismissed the Guardian Ad Litem from this Cause, but not from this Order.

This Order was rendered in open court on the 23 of August, 1991 and is effective as of that date.

The Court ORDERS the Clerk of this Court to prepare certified copies of this order for counsel and to convey immediately by telecopier the contents of this written order.

This order prohibits relator from discussing the pending litigation with anyone, ever, except in a hearing before the court, even though she has been dismissed as guardian ad litem in the case. The order even prohibits the parties from communicating with each other, although they do not complain of this prohibition here. The order is, in its own words, "an extraordinary remedy".

Relator complains that the gag orders infringe unlawfully upon her right to freedom of speech under the First Amendment to the United States Constitution. The United States Supreme Court has had several occasions to consider the validity of gag orders in criminal cases. Gentile v. State Bar of Nevada, ___ U.S. ____, 111 S.Ct. 2720, 115 L.Ed.2d 888 (1991); Nebraska Press Ass'n v. Stuart, 427 U.S. 539, 96 S.Ct. 2791, 49 L.Ed.2d 683 (1976); Sheppard v. Maxwell, 384 U.S. 333, 86 S.Ct. 1507, 16 L.Ed.2d 600 (1966); In re Sawyer, 360 U.S. 622, 79 S.Ct. 1376, 3 L.Ed.2d 1473 (1959). The principal authority applying the First Amendment to gag orders on the participants in a civil case, however, is Bernard v. Gulf Oil Co., 619 F.2d 459, 467 (5th Cir. 1980) (en banc), aff'd on other grounds, 452 U.S. 89, 101 S.Ct. 2193, 68 L.Ed.2d 693 (1981), in which the court held that an order restricting the named plaintiffs in a class action and their attorneys from communicating freely with prospective class members violated the First Amendment. The court determined that such gag orders are prior restraints upon free speech. Id. at 467. See also Rodgers v. U.S. Steel Corp., 508 F.2d 152, 162-63 (3d Cir.), cert. denied, 423 U.S. 832, 96 S.Ct. 54, 46 L.Ed.2d 50 (1975).

Prior restraints on freedom of speech have long been disfavored in American law. Near v. Minnesota, [ 283 U.S. 697, 51 S.Ct. 625, 75 L.Ed. 1357 (1931) ]. While a prior restraint is not unconstitutional per se, there is a heavy presumption against its constitutionality. Southeastern Promotions, Ltd. v. Conrad, [ 420 U.S. 546, 558-59, 95 S.Ct. 1239, 1246-47, 43 L.Ed.2d 448 (1975) ]; Organization for a Better Austin v. Keefe, [ 402 U.S. 415, 419, 91 S.Ct. 1575, 1577, 29 L.Ed.2d 1 (1971) ].

Bernard, 619 F.2d at 467. Generally, "prior restraints on speech and publication are the most serious and the least tolerable infringement on First Amendment rights." Nebraska Press, 427 U.S. at 559, 96 S.Ct. at 2803 (1976).

For a prior restraint to violate the First Amendment, it must prohibit protected activity. "The First Amendment is not absolute, and `the protection even as to previous restraint is not absolutely unlimited.'" Bernard, 619 F.2d at 471 (quoting Near, 283 U.S. at 716, 51 S.Ct. at 631); accord Nebraska Press, 427 U.S. at 570, 96 S.Ct. at 2808; Times Film Corp. v. Chicago, 365 U.S. 43, 47, 81 S.Ct. 391, 393, 5 L.Ed.2d 403 (1961); Kingsley Books, Inc. v. Brown, 354 U.S. 436, 441, 77 S.Ct. 1325, 1327-28, 1 L.Ed.2d 1469 (1957). Exceptions include speech that is obscene, seditious or extremely provocative. Nebraska Press, 427 U.S. at 590, 96 S.Ct. at 2817 (Brennan, J., concurring). It is unnecessary to decide whether there is some activity prohibited by the gag orders in this case which is not protected by the First Amendment, such as, perhaps, intimidation of the minor plaintiffs. These orders sweep far more broadly, prohibiting relator from speaking at all on any matter concerning the litigation. Beyond question, her First Amendment rights are affected.

According to Bernard, a gag order is permitted by the First Amendment only if it meets three conditions. First, "`before a prior restraint may be imposed by a judge, even in the interest of assuring a fair trial, there must be "an imminent, not merely a likely, threat to the administration of justice. The danger must not be remote or even probable; it must immediately imperil."'" Bernard, 619 F.2d at 474 (quoting United States v. Columbia Broadcasting System, Inc., 497 F.2d 102, 104 (5th Cir. 1974) (quoting Craig v. Harney, 331 U.S. 367, 376, 67 S.Ct. 1249, 1255, 91 L.Ed. 1546 (1947))). "In general, a prior restraint may be justified only if the expression sought to be restrained `surely [will] result in direct, immediate, and irreparable damage.'" Bernard, 619 F.2d at 473 (quoting International Soc'y for Krishna Consciousness v. Eaves, 601 F.2d 809, 833 (5th Cir. 1979) (quoting New York Times Co. v. United States, 403 U.S. 713, 730, 91 S.Ct. 2140, 2149, 29 L.Ed.2d 822 (1971) (Stewart, J., joined by White, J., concurring))). Even "`the interest of the judiciary in the proper administration of justice does not authorize any blanket exception to the first amendment.'" Bernard, 619 F.2d at 467 n. 8 (quoting Rodgers, 508 F.2d at 163); see Wood v. Georgia, 370 U.S. 375, 395, 82 S.Ct. 1364, 1375, 8 L.Ed.2d 569 (1962); Craig, 331 U.S. at 378, 67 S.Ct. at 1256; Pennekamp v. Florida, 328 U.S. 331, 347, 349-50, 66 S.Ct. 1029, 1037, 1038-39, 90 L.Ed. 1295 (1946); Bridges v. California, 314 U.S. 252, 272-73, 62 S.Ct. 190, 198-99, 86 L.Ed. 192 (1941). Second, a valid prior restraint "must not sweep too broadly. Rather it `must be narrowly drawn and cannot be upheld if reasonable alternatives are available having a lesser impact on First Amendment freedoms.'" Bernard, 619 F.2d at 476 (quoting CBS, Inc. v. Young, 522 F.2d 234, 238 (6th Cir. 1975)); see also Nebraska Press, 427 U.S. at 562-69, 96 S.Ct. at 2804-08; Carroll v. Commissioners of Princess Anne, 393 U.S. 175, 183-84, 89 S.Ct. 347, 352-53, 21 L.Ed.2d 325 (1968). Third, "the restraint `must have been accomplished with procedural safeguards that reduce the danger of suppressing constitutionally protected speech.'" Bernard, 619 F.2d at 477 (quoting Southeastern Promotions, 420 U.S. at 559, 95 S.Ct. at 1246). These safeguards include evidence and findings. See Nebraska Press, 427 U.S. at 564, 96 S.Ct. at 2805.

The district court's gag orders do not meet First Amendment standards. The orders were not necessitated by any imminent threat imperiling the administration of justice. The meager record before us shows that the district court was concerned that conflicts among the various participants in this litigation were resulting in misstatements and misunderstandings, jeopardizing a proposed settlement and what the court considered to be the best interests of the minor plaintiffs. The existence of those conflicts was apparent from oral argument before this Court. Relator stated quite clearly that as former guardian ad litem she had and still has very strong views about the children's interests. The district court clearly believed that the plaintiffs needed to be protected from relator's insistence on expressing her views, and the other parties appear to share the court's view, joining in defense of the gag orders which they are themselves subject to. As far removed from the conduct of the litigation as we are, it is difficult to evaluate the district court's concerns. Assuming, however, that relator was every bit the threat to her previous wards that the district court considered her to be, that threat did not impinge so imminently upon the administration of justice as to satisfy the first condition of Bernard.

Moreover, the orders were overly broad and were not the only reasonable alternative for addressing the problems the district court confronted. The district court prohibited relator from talking with anyone about the case under any circumstances except in the course of proceedings. This order is far more expansive than the order struck down in Bernard; it does not merely limit relator's communications about the case, it prohibits them altogether outside the courtroom. Although the district court in this case met with the parties in an effort to dispel confusion, and although it cautioned relator against causing further conflicts and misunderstandings, it did not reasonably exhaust these efforts or explore the use of disciplinary measures or sanctions against relator before drastically restricting her fundamental rights. Removing relator as guardian ad litem might alone have alleviated the conflicts. The district court neither exhausted reasonable alternative measures nor limited its prohibition to what was necessary to accomplish its purposes. Thus the gag orders cannot meet Bernard's second condition.

Finally, the district court did not follow procedures that would safeguard against an unwarranted infringement of relator's First Amendment rights. The district court acted on its own, without motion or argument from the parties. Although the district court conducted an evidentiary hearing at some point prior to issuing its first order, the record does not reflect whether any of that evidence pertained to the necessity and scope of a gag order. The orders come to this Court supported by a single, general finding recited in the written order. These procedures do not comport with those required by Bernard's third condition.

Bernard does not hold that participants in civil court proceedings have a boundless constitutional right to speak extrajudicially about the litigation, or that a court is powerless to limit such speech in all circumstances. It does, however, set a high standard for any such limitation in recognition of the importance of free speech rights of attorneys and litigants. The issuance of gag orders in these circumstances did not meet this standard and therefore was an unlawful infringement upon relator's constitutionally protected freedom of speech and a clear abuse of discretion. Relator has no right to appeal these interlocutory orders until final judgment is rendered and meanwhile must suffer their irreversible and irrecompensable effects. She is therefore entitled to mandamus relief, which we have already granted, directing the district court to vacate the gag orders.

B

The Court reaches this result but not by the well-traveled road of First Amendment jurisprudence. It insists instead on traversing the largely uncharted terrain of article I, section 8 of the Texas Constitution.

1

I say "insists" because the idea that this case should be decided on state constitutional grounds alone did not originate with the parties but with one Member of this Court. Article I, section 8 was not even mentioned in this case until relator's counsel alluded to it in passing in oral argument, after which the following colloquy occurred:

JUSTICE DOGGETT: Are you asserting — you made reference to the Texas Constitution earlier — are you asserting free speech rights under the Texas Constitution as well as the U.S. Constitution?

RELATOR'S COUNSEL: That is correct, Your Honor. I read article I, section 8 of the Texas Constitution as going beyond the First Amendment.

JUSTICE DOGGETT: Have you briefed that and cited us any authority on the broader protections afforded by the Texas Constitution?

RELATOR'S COUNSEL: No, we haven't, Your Honor. I entered this case after the briefing was terminated.

JUSTICE DOGGETT: Are you interested in filing any supplemental briefing on that issue?

RELATOR'S COUNSEL: We would welcome the opportunity to file supplemental briefing with this Court on the First Amendment issues.

There was no further discussion of article I, section 8 during oral argument. Relator subsequently filed a supplemental brief arguing that the gag orders were not permitted by the First Amendment, and secondarily, that article I, section 8 provides an independent basis for overturning them. Respondent never replied to this brief.

Thus, in considering the applicability of article I, section 8 to this case, the Court has the benefit of no argument and approximately four pages of briefing which relator filed at the invitation of one Justice. The Court may choose to rest a significant decision on grounds which the parties have not fully presented, but there are always risks in doing so, and the more important the issue, the greater the risks. The validity of prior restraints under article I, section 8, independent of First Amendment law and any law in other states, is a very important issue. Had it been fully presented by the parties, and had our understanding of this provision of our constitution grown and developed over time, the issue might be addressed with more assurance. Today, however, Texas prior restraint law is born a teenager, a process as remarkable as it is frightful.

2

Article I, section 8 of the Texas Constitution states in pertinent part:

Every person shall be at liberty to speak, write or publish his opinions on any subject, being responsible for the abuse of that privilege; and no law shall ever be passed curtailing the liberty of speech or of the press.

This provision in our state Bill of Rights, like the First Amendment to the United States Constitution and similar provisions in the constitutions of other states, enshrines and protects a fundamental right long treasured by the people of this nation, the right of free speech. That right, however, is not absolute, as we long ago learned in our thinking about the First Amendment. Justice Holmes' classic example, familiar to lawyers and non-lawyers alike, is that one does not have the right of "falsely shouting fire in a theatre and causing a panic." Schenck v. United States, 249 U.S. 47, 52, 39 S.Ct. 247, 249, 63 L.Ed. 470 (1919). Free speech can be abused, as article I, section 8 expressly recognizes, and responsibility for that abuse is not only consistent with protecting the freedom, it is part of the freedom itself.

Freedom and responsibility have a symbiotic relationship: they are part of one another, yet in tension. So here, relator contends that she should be free to communicate with the parties to this litigation, and the district court counters that relator's freedom should be restricted because of her responsibility not to cause misunderstandings which threaten the best interests of the minor plaintiffs. Article I, section 8 provides principles for resolving this dispute, but it does not prescribe the resolution. It falls to the Court to determine how these governing principles apply in specific situations. To make this determination we ordinarily look to such things as the language of the constitutional provision itself, its purpose, the historical context in which it was written, the intentions of the framers, the application in prior judicial decisions, the relation of the provision to the law as a whole, the understanding of other branches of government, the law in other jurisdictions, state and federal, constitutional and legal theory, and fundamental values including justice and social policy. See, e.g., Brown v. Meyer, 787 S.W.2d 42, 45 (Tex. 1990); Edgewood Indep. Sch. Dist. v. Kirby, 777 S.W.2d 391, 394 (Tex. 1989); State v. Project Principle, Inc., 724 S.W.2d 387, 390-91 (Tex. 1987); Spring Branch Indep. Sch. Dist. v. Stamos, 695 S.W.2d 556, 559-62 (Tex. 1985); City of El Paso v. El Paso Community College Dist., 729 S.W.2d 296, 298 (Tex. 1986); Tarrant County v. Ashmore, 635 S.W.2d 417, 420-23 (Tex. 1982); Gragg v. Cayuga Indep. Sch. Dist., 539 S.W.2d 861, 865-66 (Tex.), appeal dismissed, 429 U.S. 973, 97 S.Ct. 478, 50 L.Ed.2d 581 (1976); Ex parte Werblud, 536 S.W.2d 542, 544-48 (Tex. 1976); Harris v. City of Fort Worth, 142 Tex. 600, 180 S.W.2d 131, 133 (1944); Jones v. Ross, 141 Tex. 415, 173 S.W.2d 1022, 1024 (1943); Collingsworth County v. Allred, 120 Tex. 473, 40 S.W.2d 13, 15 (1931); see also Dir. of the Dep't of Agric. and Env't v. Printing Indus. Ass'n, 600 S.W.2d 264, 267 (Tex. 1980); Travelers' Ins. Co. v. Marshall, 124 Tex. 45, 76 S.W.2d 1007, 1012-25 (1934); Mellinger v. City of Houston, 68 Tex. 37, 3 S.W. 249, 252-54 (1887).

In applying article I, section 8 to the gag orders here, the Court attempts to follow this process as it ordinarily would, except that, consistent with its new approach, it expressly refuses to consider whether the orders violate the First Amendment. The Court tries to prove that article I, section 8 can be construed and applied completely independently of the First Amendment. To do this, the Court examines the history of article I, section 8, its text, and caselaw construing it. Its analysis of each of these three areas is seriously deficient, as I shall show before reviewing the Court's conclusions.

3

The Court recites a little history: that Stephen F. Austin was jailed and Lorenzo de Zavala hunted down for being outspoken; that provisions guaranteeing free speech were included in the first proposed Texas Constitution in 1833 and the constitutions of 1836, 1861, 1866, 1869 and 1876; that framers of these constitutions represented a "heterogenous miscellany of opinions" different from the framers of the U.S. Constitution; and that after debate at times vigorous on topics including secession, these framers rejected the free speech provision of the Tennessee Constitution, as well as a provision regarding speech injurious of female reputation libelous without regard to its truth and a provision conditioning free speech on good motives. The provision proposed for the Texas Constitution in 1833 stated:

The free communication of thoughts and opinion, is one of the inviolable rights of man; and every person may freely speak, write, print, and publish, on any subject, being responsible for the abuse of that liberty. . . .

Proposed Constitution for the State of Texas (1833) art. 16, reprinted in Documents of Texas History 80 (Ernest Wallace ed. 1963). This earlier proposal was virtually identical to the Tennessee provision rejected in 1876 by the Texas framers, except only that the latter was limited to citizens. Tennessee's Constitution, first adopted in 1796, provides:

The free communication of thoughts and opinions is one of the invaluable rights of man; and every citizen may freely speak, write, and print on any subject, being responsible for the abuse of that liberty.

TENN. CONST. art. XI, Sec. 19 (1796) (reprinted in BENJAMIN PERLEY POORE, THE FEDERAL AND STATE CONSTITUTIONS, COLONIAL CHARTERS AND OTHER ORGANIC LAWS OF THE UNITED STATES, Vol. 1 2 (2d ed. 1878)). The other two provisions which the Texas framers rejected would have imposed significant limitations on the right of free speech.

These interesting but casual historical notes, which the Court optimistically calls a "rather extensive historical discussion", ante, at 22, indicate that free speech has always been very important in Texas, something I have never supposed was in dispute, but say nothing about how article I, section 8 applies to gag orders, or why. There is a reason for the Court's shallow approach: it does not intend to be bound to any historical interpretation. Historical analysis is "only a starting point", the Court says; "[i]n no way must our understanding of its guarantees be frozen in the past." Ante, at 19. In other words, the Court does not accept the premise that the intent of the framers of the constitution governs its future construction. "[W]e use history," the Court explains, "to assist in an understanding of the generalities and ambiguities sometimes present in a constitution." Ante, at 19. If this is true, what "generalities and ambiguities" in article I, section 8 has the Court clarified by its observations about Stephen F. Austin and Lorenzo de Zavala, or any other aspect of its "rather extensive" historical discussion? I cannot find one.

4

The Court also examines the text of article I, section 8, pointing out the inescapable facts that the language is not identical to that of the First Amendment, that it is stated partly in the affirmative ("Every person shall be at liberty to speak") rather than entirely in the negative ("Congress shall make no law"), and that it appears toward the front of the constitution rather than at the end. These facts prove conclusively that the language of article I, section 8 is different from the First Amendment, something that is plain as day. The issue, however, is not whether there are differences, but what, if anything, those differences mean, and the Court's observations shed no light on this issue. The Court cannot substantiate its claim that the framers of the Texas Constitution "explicitly rejected" verbatim adoption of the First Amendment of the U.S. Constitution, but even if it could, we would not know what the framers intended in so doing.

The Court's contention that the language of article I, section 8 is "different" and "highly distinct" is misleading. Texas, like most states, appears to have derived its constitutional provision protecting the right to free speech from Blackstone, who articulated that right as follows: "Every freeman has an undoubted right to lay what sentiments he pleases before the public; to forbid this, is to destroy the freedom of the press: but if he publishes what is improper, mischievous, or illegal, he must take the consequences of his own temerity." 4 WILLIAM BLACKSTONE, COMMENTARIES 152. By 1833, when a constitution was first proposed for Texas, 15 of the 24 states then in the United States had constitutional provisions protecting free speech in words similar to Blackstone's. The language proposed for Texas' constitution in 1833, and the language incorporated in the 1876 constitution, are obviously related to Blackstone's formulation of the common law, and are strikingly similar to the provisions of what were then almost two-thirds of the state constitutions except in one respect: the language of our 1876 constitution, unlike the language proposed in 1833, extended freedom of speech to persons instead of citizens. Even in this respect, however, Texas was neither unique nor first: Missouri had already adopted such a provision. Supra note 8.

The close relationship between the free speech guarantees in the constitutions of Texas and many other states should not be surprising. The idea that freedom of speech is a fundamental right is not unique to Texas or any other state, but one inherent in our political structure and shared generally by the people of this nation. While the right has been described in somewhat different words at different times and places, the basic ideas certainly transcend state lines. Free speech is a national idea, not only a Texas idea. The Court's attempt to distinguish Texas free speech as significantly different from First Amendment free speech — and presumably also from New York free speech or California free speech — is not supported by the texts of the various guarantees.

5

The only serious effort the Court makes to determine how article I, section 8 should apply to gag orders is by examining our own precedents, and this effort, though flawed, is exhaustive. For it should be noted that in more than 150 years article I, section 8 has been mentioned in this Court's opinions in only 19 cases, and twice the reference was in a separate opinion. In only five of these 19 cases has this Court written more than a few words about article I, section 8, and only three of those five involved prior restraints. Two of the three prior restraint cases relied entirely on the third, Ex parte Tucker, 110 Tex. 335, 220 S.W. 75 (1920). This is the totality of our prior restraint jurisprudence under the Texas Constitution. Tucker, decided before the First Amendment was applied to the states through the Fourteenth Amendment in Gitlow v. New York, 268 U.S. 652, 45 S.Ct. 625, 69 L.Ed. 1138 (1925), held that an injunction prohibiting union employees from "`vilifying, abusing, or using opprobrious epithets to'" other employees violated article I, section 8. Tucker, 220 S.W. at 75. The Court cites Tucker and quotes from it extensively as an example of the independence of article I, section 8 from the First Amendment. But the Court does not quote the following:

The experience of the English nation and some of the American colonies under the tyranny of such systems is the reason this provision in the Bill of Rights [article I, section 8] is one common to the Constitutions of the American States, and for its incorporation, in like words, in the First Amendment to the Federal Constitution.

Id. at 76 (emphasis added). Tucker did not ignore the First Amendment in its analysis of article I, section 8, as the Court has, nor did it differentiate the two provisions, as the Court attempts to do today. Rather, it linked them in two respects: they were founded on common experience, that of the English nation and the American colonies, and they were framed in "like words".

The other two cases which utilize article I, section 8 to invalidate prior restraints rely entirely upon Tucker. In Dallas General Drivers v. Wamix, Inc., 156 Tex. 408, 295 S.W.2d 873, 879 (Tex. 1956), the Court dissolved an injunction prohibiting striking employees from "us[ing] . . . insulting, threatening and indecent language" toward non-striking employees "without prejudice to the right of the trial court to reinstate it if future conduct of the [striking employees] should authorize it." Thus, the Court invalidated an injunction as a prior restraint but did not preclude the trial court from reissuing it if the circumstances warranted. This is certainly not a very broad reading of article I, section 8, and may well be narrower than the First Amendment. In Hajek v. Bill Mowbray Motors, Inc., 647 S.W.2d 253, 255 (Tex. 1983) (per curiam), the Court dissolved an injunction prohibiting an owner from driving his car on which he had prominently labeled as a "lemon". Although the Court did not refer to the First Amendment, it also did not state that article I, section 8 afforded broader protection of speech.

Thus, all three of this Court's prior restraint cases have at least assumed a congruence between article I, section 8 and the First Amendment. The other two decisions of this Court in which article I, section 8 is discussed at all are Casso v. Brand, 776 S.W.2d 551, 556 (Tex. 1989), and O'Quinn v. State Bar of Texas, 763 S.W.2d 397, 402 (Tex. 1988). The Court quotes Casso as stating that "our state free speech guarantee may be broader than the corresponding federal guarantee". Ante, at 9. What Casso actually said was:

While we have recently recognized the possibility that our state free speech guarantee may be broader than the corresponding federal guarantee, see O'Quinn v. State Bar, 763 S.W.2d 397, 402 (Tex. 1988), that broader protection, if any, cannot come at the expense of a defamation claimant's right to redress. Unlike the United States Constitution, which contains no explicit guarantee of the right to sue for defamation, the Texas Constitution expressly protects the bringing of reputational torts. . . .

These provisions must be given effect. While we may on occasion grant protections to defamation defendants beyond those required in the United States Constitution, as we have today in requiring public official and public figure plaintiffs to prove their actions against private defendants under the New York Times [Co. v. Sullivan, 376 U.S. 254, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964) ] standard, we have based those decisions on common law, not constitutional grounds.

Casso, 776 S.W.2d at 556 (emphasis added). The Court's quote is somewhat misleading. While Casso admits the "possibility" that article I, section 8 may contain a broader guarantee of free speech than the First Amendment, it also recognizes that the state provision expressly imposes responsibility for abusive speech, something which the First Amendment does not do. On the whole, Casso not only fails to lend the Court's position support, it undercuts it. The Court also quotes O'Quinn as stating, "`Texas' free speech right [has been characterized] as broader than its federal equivalent'". Ante, at 8. What the Court actually said was: " One commentator has characterized Texas' free speech right as being broader than its federal equivalent. . . ." O'Quinn, 763 S.W.2d at 402 (emphasis added). O'Quinn does not claim that a broader view of article I, section 8 is as widely held as the Court would suggest. Nor did the Court in O'Quinn embrace the view, adding: "We need not decide at this time whether Texas' guarantee of free speech affords greater protection than its corresponding federal rights. . . ." Id.

These five decisions, Tucker, Dallas General, Hajek, Casso and O'Quinn, constitute the entirety of our article I, section 8 caselaw. As if they were hardly sufficient foundation for a rule governing gag orders, the Court also cites two court of appeals decisions, Amalgamated Meat Cut. v. Carl's Meat Provision Co., 475 S.W.2d 300, 304 (Tex.Civ.App. — Beaumont 1971, writ dism'd w.o.j.), and Pirmantgen v. Feminelli, 745 S.W.2d 576, 578 (Tex.App. — Corpus Christi 1988, no writ), cases based upon both the federal and state provisions. Contrary to the Court's new approach in this case, both these cases rely in part upon the First Amendment. The last time this Court addressed a prior restraint was in Iranian Muslim Org. v. City of San Antonio, 615 S.W.2d 202 (Tex. 1981), where we invalidated the restraint based solely upon the First Amendment without alluding to article I, section 8. The Court does not cite Iranian. Two court of appeals decisions which the Court also does not cite expressly hold that "Texas constitutional provisions guaranteeing freedom of expression and assembly are coextensive with the corresponding federal guarantees". Puckett v. State, 801 S.W.2d 188, 192 (Tex.App.Houston [14th Dist.] 1990), cert. denied, ___ U.S. ___, 112 S.Ct. 606, 116 L.Ed.2d 629 (1991); Reed v. State, 762 S.W.2d 640, 644 (Tex.App. — Texarkana 1988, pet. ref'd), cert. denied, 493 U.S. 822, 110 S.Ct. 81, 107 L.Ed.2d 47 (1989). Reed adds, "and we will apply the same analysis and principles of construction in interpreting them." Id.

Finally, the Court refers to two opinions of our Court of Criminal Appeals, both of which held that newspaper publishers could not be held in contempt for publishing criminal trial testimony in violation of a court order. Both decisions were based upon article I, section 8. However, Ex parte Foster, 44 Tex.Crim. R., 71 S.W. 593 (App. 1903), disapproved on other grounds in Ex parte Winfree, 153 Tex. 12, 263 S.W.2d 154, 158 (1953), like Tucker, was decided before the First Amendment was applied to the states. And Ex parte McCormick, 129 Tex.Crim. R., 88 S.W.2d 104, 106 (App. 1935), relied upon both federal and state caselaw, and observed that the guaranty of article I, section 8 is also embodied in state constitutions and in the First Amendment. Neither case follows the Court's approach in this case of ignoring First Amendment law, and neither distinguishes article I, section 8 from the First Amendment.

The truth of the matter is that all our prior caselaw either assumes a close identity between the First Amendment and article I, section 8, or is silent on the subject. Not before today has this Court insisted that the two provisions are different in substance, and so much so that we should not even consider the former in construing the latter.

6

It is important to note that the Court does not tie its analysis of the gag orders in this case to the history of article I, section 8, or to its text, or to any prior caselaw. Rather, it states: "Since the dimensions of our constitutionally guaranteed liberties are continually evolving, today we build on our prior decisions by affirming that a prior restraint on expression is presumptively unconstitutional." Ante, at 10. By "continually evolving", the Court means that it is free to construe our constitution unconstrained by its history or any prior construction. "With this concept in mind," the Court adopts a test for determining whether gag orders are valid under article I, section 8, the basis for which cannot be found in any kind of precedent that the Court recognizes. That test is as follows:

a gag order in civil judicial proceedings will withstand constitutional scrutiny only where there are specific findings supported by evidence that (1) an imminent and irreparable harm to the judicial process will deprive litigants of a just resolution of their dispute, and (2) the judicial action represents the least restrictive means to prevent that harm.

Ante, at 10.

In deciding upon this test, the Court states that its analysis has been assisted by federal cases, and it cites two: Nebraska Press and Bernard. The Court criticizes both, although they both struck down the gag orders involved. Nebraska Press, it says, is a "fact specific" and "splintered" decision, ante, at 10, which is "too permissive", ante, at 11, and "does not . . . sufficiently protect the rights of free expression that we believe that the fundamental law of our state secures", ante, at 10. The Court does not explain how Nebraska Press is any more fact specific than this case, or how it leaves fundamental rights unguarded. Bernard, the Court says, was decided "in the context of Rule 23(d) of the Federal Rules of Civil Procedure", ante, at 10 n. 16, "on a very splintered vote, and thereafter disregarded by the United States Supreme Court", ante, at 21, showing "[i]f anything, the admittedly unsettled nature of the federal law", ante, at 10 n. 16. The truth is, that while Bernard involved federal rule 23, it was decided on First Amendment grounds; that the "very splintered vote" was thirteen judges for the court's opinion, eight concurring, and one dissenting — not too different from our vote in this case; and that the U.S. Supreme Court affirmed the judgment of the Fifth Circuit without reaching the First Amendment issues. As for whether federal law is "admittedly unsettled", Bernard simply assembles and restates the holdings of a number of U.S. Supreme Court decisions in fashioning a test for gag orders.

As flawed as the Court considers Nebraska Press and Bernard to be, it is difficult to conceive how the Court can state a test for gag orders under article I, section 8 that is identical to Bernard's First Amendment standards. The presumption against the constitutionality of gag orders is the same. The first element of the Court's test — "an imminent and irreparable harm to the judicial process [that] will deprive litigants of a just resolution of their dispute" — is the same as Bernard's — "an imminent, not merely a likely, threat to the administration of justice", 619 F.2d at 474 (quoting U.S. v. Columbia Broadcasting System, Inc., 497 F.2d 102, 104 (5th Cir. 1974)), that "surely [will] result in direct, immediate, and irreparable damage", 619 F.2d at 473 (quoting International Soc'y for Krishna Consciousness, 601 F.2d at 833 (quoting New York Times, 403 U.S. at 730, 91 S.Ct. at 2149)). The second element of the Court's test — "the judicial action represents the least restrictive means to prevent that harm" — is also the same as Bernard's — the gag order "`must be narrowly drawn and cannot be upheld if reasonable alternatives are available having a lesser impact on First Amendment freedoms' ", 619 F.2d at 476 (quoting CBS, 522 F.2d at 238). The Court's requirement that there be specific findings supported by evidence is the same as Bernard's third condition that there be "procedural safeguards" including evidence and findings, 619 F.2d at 477 (quoting Southeastern Promotions, 420 U.S. at 559, 95 S.Ct. at 1246).

The identity between First Amendment standards for reviewing gag orders and the Court's new standards under article I, section 8 is a very remarkable result considering that the Court's opinion calls for "an independent standard under the Texas Constitution." Ante, at 11. After all, the two constitutional provisions have different authors, different words, and different histories, and according to the Court, they have been and should be treated separately. Although the Court professes not even to have considered the validity of the gag orders under the First Amendment, the test it announces for applying the unique and distinctive language of article I, section 8 turns out to be identical to the First Amendment test. This can happen, the Court says: "independent interpretation [does not] necessarily yield a different result". Ante, at 22. The Court explains: "Our investigation may reveal federal authority so complete, so well reasoned, and so consistent with the provisions of the Texas Constitution in protecting the individual liberties that we reach the same conclusion." Id. It might, of course, but here it does not; that is, the Court rejects the only two federal cases it cites relating to a test for gag orders as being incomplete, poorly reasoned, and not fully protective of state constitutional free speech rights. The Court has explained how it could reach the same conclusion as federal cases which it considered well-reasoned and authoritative; it has not explained how it could reach the same conclusion as federal cases which it rejects.

If the two constitutional provisions are really as different as the Court insists they are, it is a remarkable coincidence that the standards for applying them to gag orders happen to be identical. But if the standards really are the same, then there is no practical difference in the two provisions, contrary to the Court's insistence. Thus, the Court's analysis strains credulity and finally disproves its own thesis.

7

Without reference to the First Amendment, or to the hundreds of cases construing it, this Court attempts to formulate independent standards for applying the guarantee of free speech in article I, section 8 of the Texas Constitution to one form of prior restraints, gag orders like the ones in this case. The task is daunting, even though the law the Court attempts to create on its own is but a small part of what Americans have come to understand as freedom of speech. And in the end the Court fails in its efforts. Barely managing to cobble together a few fragments of history, obvious truisms about the constitutional language, exaggerated claims about its distinctiveness, and phrases taken out of context from a few of our cases, the Court produces a test that is identical to more fully developed First Amendment standards. The Court achieves the very end it sought to avoid — adoption of First Amendment standards — without admitting it.

Why? If state and federal constitutional law conflicted, or if federal law were undeveloped or nonexistent, an effort to expound state law might be productive. But these circumstances are not present here. The Court's effort in this case is like creating a new language in order to write a novel: it is possible to do it, but unnecessary when author and readers already share a common language. And one cannot help being skeptical of an author who claims to have written a book in a new language when the new language sounds a lot like English and the book reads a lot like MOBY DICK.

C

Most of the Court's opinion today is devoted to a defense of its new method of constitutional analysis which examines the state constitution first, and if a right is found to be protected, never reaches the federal constitutional question. The Court derives this approach from developments in other jurisdictions and our own caselaw. Neither supports the Court's new methodology.

1

The Court claims that its new method of constitutional analysis is part of a "trend" that "has met with broad approval" and has been endorsed overwhelmingly by state and federal courts as well as commentators throughout the nation. Ante, at 12. These claims are greatly exaggerated. Certainly, there are a number of courts and commentators who have advocated an approach to state courts' decisions of constitutional issues like the one the Court uses today. But the thinking on the subject is not all one way, a fact which the Court attempts to minimize. The truth is that a substantial body of legal commentators disagrees that an approach like the Court uses is proper or even workable. Most of the cases the Court cites from other jurisdictions are not really as favorable as it suggests. And the Court's assertion that "federal courts have encouraged state courts to embark upon independent analysis of their own constitutions", ante, at 15, is not only unsupported by authority but highly improbable. (Why would federal courts take it upon themselves either to encourage or discourage state courts in applying state constitutions?)

As the Court notes, many of the authorities I have cited support judicial reliance upon state constitutions. So do I. Contrary to the Court's assertions, I do not argue that state constitutions should be ignored, or that federal law always controls their construction. I contend only that federal constitutional law should be considered when a state constitutional provision is applied. The Court's new analytical method treats the resource of federal law with marked ambivalence. On the one hand, the Court includes federal law among the considerations its new method is to use: "Within the context of such an analysis, a state court can benefit from the insights of well-reasoned and developed federal jurisprudence, but is not compelled to reach identical results." Ante, at 13. On the other hand, the Court avoids any useful reference to federal law in this case. The Court claims that it "quite obviously" does not disregard federal law, yet it ignores Bernard and the United States Supreme Court decisions on which it relies. The Court says First Amendment jurisprudence is "not irrelevant", but it also says that it "need not consider whether the United States Constitution has also been violated." Ante, at 11. In sum, the Court says federal law can be useful in applying the state constitution, but the Court does not use it in this case.

To portray its new methodology in a more favorable light, the Court contrasts it with a misstatement of this opinion:

Rejecting our careful and detailed analysis of the development and interpretation of article one, section eight, the concurrence advances an alternative — Texas judges should follow, but never lead, federal jurisprudence. . . . [T]he Texas judge should never diverge from the path taken by the federal judiciary. . . . [Nothing] should obscure the obligation of adherence to federal authority.

The Court's attempt to focus constitutional analysis on state law to the exclusion of federal law is at odds with itself. If the Court acknowledges that the test for gag orders under federal law is identical to the test it adopts under article I, section 8, it can hardly claim that it has arrived at this test independently. If the Court refuses to acknowledge federal law, then it assumes the difficult task of constructing a state test from almost no precedent, only to arrive at the very conclusion federal law dictates. The Court's approach insists upon looking for differences between the state and federal constitutions when none can be found.

One extensive article surveying the thinking concerning "new federalism" explains some of the deficiencies in the Court's approach:

Several observers of recent state constitutional activism have argued that state constitutions should be regarded as the primary sources of individual rights and liberties and that state courts should interpret state constitutions without reference to "all the old, familiar shorthand" of federal constitutional law. According to this "primacy" model, the state court should consider assertions of federal constitutional rights only after all claims resting on state law have failed to provide the requested protections. The assumption underlying this model is that the states are the primary sovereigns and that state constitutions are the basic charters of individual liberties and of the limits of governmental authority. In this model, federal law, including the fourteenth amendment, provides only limited constraints on state autonomy.

The failing of the primacy model is that this assumption no longer resembles reality. Nor does it reflect the fact that litigants typically present state constitutional issues only when they expect an unfavorable federal constitutional result. Federal assumption of the dominant role in the federal system — and particularly

in the protection of individual rights — has rendered the primacy model obsolete. When federal protections are extensive and well articulated, state court decisionmaking that eschews consideration of, or reliance on, federal doctrine not only will often be an inefficient route to an inevitable result, but also will lack the cogency that a reasoned reaction to the federal view could provide, particularly when parallel federal issues have been exhaustively discussed by the Supreme Court and commentators. In a community that perceives the Supreme Court to be the primary interpreter of constitutional rights, reliance on Supreme Court reasoning can help to legitimate state constitutional decisions that build on the federal base. When a state court diverges from the federal view, a reasoned explanation of the divergence may be necessary if the decision is to command respect.

For state constitutional law to assume a realistic role, state courts must acknowledge the dominance of federal law and focus directly on the gap-filling potential of state constitutions. This interstitial role recognizes federal doctrine as a settled floor of rights and asks whether and how to criticize, amplify, or supplement this doctrine to yield more extensive constitutional protections. The state court's role is not to construct a complete system of fundamental rights from the ground up. [Footnotes omitted.]

Developments in the Law — The Interpretation of State Constitutional Rights, 95 HARV.L.REV. at 1356-1358; see also Stewart G. Pollack, State Constitutions as Separate Sources of Fundamental Rights, 35 RUTGERS L.REV. 707, 718.

It cannot be denied that there are rights protected by state constitutions that extend beyond those guaranteed by the United States Constitution. Many state constitutional provisions simply have no federal analogue. Three of the most important decisions this Court has ever issued were based upon such provisions. Carrollton-Farmers Branch Indep. Sch. Dist. v. Edgewood Indep. Sch. Dist., 826 S.W.2d 489 (Tex. 1992); Edgewood Indep. Sch. Dist. v. Kirby, 804 S.W.2d 491 (Tex. 1991); Edgewood Indep. Sch. Dist. v. Kirby, 777 S.W.2d 391 (Tex. 1989). In construing state constitutional provisions which have no federal counterpart, reference to federal law is usually of little utility.

When state and federal provisions overlap or correspond, state law, as well as federal law and the law of other states, may be helpful in analyzing their proper application. To ignore all federal constitutional law in construing state constitutional provisions guaranteeing rights common to both is as wrong as ignoring state constitutional provisions altogether. If nothing else, it is inefficient to blaze a trail through the wilderness when there is a perfectly good highway there already, built at considerable expense, and well traveled. But the problems of the Court's method run even deeper. The Court does not merely ignore federal law; it rejects it. And the rejection has a disturbing tone to it. "[O]ur concerns are Texas concerns," the Court asserts, a viewpoint that cannot be very comforting to out-of-state parties litigating in Texas courts.

2

The Court's claim that Texas courts have "recognized the importance of our state constitution" for more than a century, ante, at 13, cannot be disputed. Certainly, if state courts have not recognized the importance of our state constitution, they should have. This does not mean, of course, that any Texas court has ever employed the constitutional analysis used by the Court today. If the Court's new analytical method had really been followed in Texas for 100 years, as the Court means to suggest, it would hardly need the major defense the Court attempts to provide in this case. Today's opinion is significant only because the Court's methodology has not previously been the accepted model in Texas. In five of the cases the Court cites, we applied provisions of our constitution which have no federal counterpart: In re Baby McLean, 725 S.W.2d 696, 698 (Tex. 1987) (article I, section 3a, equal rights amendment); LeCroy v. Hanlon, 713 S.W.2d 335, 341-42 (Tex. 1986) (article I, section 13, open courts); Nelson v. Krusen, 678 S.W.2d 918, 922 (Tex. 1984) (open courts); Sax v. Votteler, 648 S.W.2d 661, 664 (Tex. 1983) (open courts); Bell v. Indian Live-Stock Co., 11 S.W. 344, 345 (Tex. 1889) (article 16, section 28, protecting current wages for personal service from garnishment). Travelers' Ins. Co. v. Marshall, 124 Tex. 45, 76 S.W.2d 1007 (1934), analyzed both state and federal constitutional law, as many other cases from this Court have. Supra note 6. In Tucker, as we have noted, the Court referred to article I, section 8 and the First Amendment as "like" provisions, and Hajek relies entirely on Tucker. In Whitworth v. Bynum, 699 S.W.2d 194, 197 (Tex. 1985), the Court applied state equal protection guarantees to declare our Guest Statute unconstitutional, but drew upon federal principles of analysis. Foster was decided by the Court of Criminal Appeals at a time when the First Amendment had not clearly been applied to the states. Kemper v. State, 63 Tex.Cr.R. 138 S.W. 1025, 1044-1045 (App. 1911), cited by the Court for the proposition that even an opinion of the United States Supreme Court should be questioned by Texas courts if it improperly disregards the rights of Texans, was explicitly overruled on this very point one year later in Robertson v. State, 63 Tex.Crim. R., 142 S.W. 533, 546 (App. 1912) (holding that the right to confrontation under article I, section 10 of the state constitution shares a common heritage and interpretation with the Sixth Amendment, and deferring to the Supreme Court's interpretation that of right). And McCormick and Heitman v. State, 815 S.W.2d 681 (Tex.Crim.App. 1991), consider both federal and state law in construing constitutional rights. None of the Texas cases the Court cites uses the method it endorses.

The Court's method is especially problematic in a case involving freedom of speech. As one commentator states:

There are good reasons for the state courts to look to federal law for guidance in the first amendment area, an area in which the issues now being addressed are intricate and difficult. The state courts are wrestling with essentially the same questions as the federal courts, and all approach those issues without a clear constitutional mandate. The states would therefore be foolish to ignore the convenient resource presented by the federal experience; an argument is no less persuasive because it relies upon or quotes an argument made elsewhere. Commentators who condemn state judiciaries for referring to federal doctrine when interpreting their own charters would force an irrational chauvinism on the state courts.

Developments in the Law — The Interpretation of State Constitutional Rights, 95 HARV.L.REV. at 1419 (emphasis added).

3

One argument the Court makes for its new method — avoidance of "unnecessary" federal review — is more subtle and requires more attention. The Court contends that its approach is more efficient because if a case is decided on state constitutional grounds, the United States Supreme Court cannot review it, and thus that Court's workload is reduced and the parties' dispute more quickly resolved. As evidence of the savings to be achieved using its approach, the Court cites a total of four decisions by state courts in the past seventeen years in which review by the United States Supreme Court could have been avoided. Even if it is assumed that all four cases could have been decided on state constitutional grounds unreviewable by the Supreme Court, reducing that Court's workload by four cases in 17 years would not measurably improve its efficiency. Nor do delays in so few cases over so long a period indicate a problem of any magnitude. Efficiency is not a very compelling basis for the Court's argument.

But the Court has more serious reasons for wanting to avoid "unnecessary" federal review. The Court argues that enforcing state constitutional rights both protects values fundamental to the people of that state and commends them to the rest of the nation. The people of a state speak through their state constitution, the argument runs, and that voice should rule in the state and be heard in the nation. This argument, though true in some respects, is mostly a rhetorical appeal to state pride. More importantly, it understates the Court's goal. If a state constitutional provision has no federal counterpart, it must be given effect so long as that can be done without infringing upon the federal constitution. If state and federal constitutional provisions overlap, both must be considered; if they conflict, the state provision must give way. The undeniable fact that the people of a state are entitled to a voice in their government does not mean that the voice of the nation of which the state is a part can be ignored. The goal of the Court's methodology is not merely to augment a state's voice in national affairs, but to still the national voice in state affairs. The approach adopted by the Court is not limited to producing state participation in the national debate over fundamental issues; it seeks state autonomy.

That autonomy, the Court recognizes, is accomplished not by the state constitution itself, but by the interpretation of the constitution by the state's highest court. And this brings us to what lies at the very heart of the Court's position: the Justices of this Court, and not the United States Supreme Court, should determine the people's fundamental rights, and if our determinations are unsatisfactory, we can be replaced. By way of illustration, the Court cites Ex parte Rodriguez, 39 Tex. 706 (1873), as an unpopular decision resulting in the removal of the Justices of this Court by the Governor. This is the real sense in which federal review is "unnecessary".

There are several difficulties with the Court's position. First, it weakens the validity of constitutionalism and the rule of law. The essence of constitutionalism is that certain principles, endorsed by the people, become fundamental rules of law. How these rules apply in changing circumstances is often disputed, and the judiciary in this country has taken upon itself the ultimate responsibility of resolving those disputes, beginning with Marbury v. Madison, 5 U.S. (1 Cranch) 137, 2 L.Ed. 60 (1803). The people and the other branches of government have acquiesced in this delegation of responsibility in part, I think, because of their perception that the judiciary is bound in its construction and application of constitutional provisions by definite rules of law which preclude judges from substituting their personal policies for constitutional principles in the guise of construing those principles. Adherence to these rules is essential to the validity and the credibility of constitutionalism. For this reason, constitutional construction must be founded upon a careful construction of each provision's language, purpose, history and intent, as well as upon precedent, theory and fundamental values. If the analytical process is inadequate or flawed, the result may appear to be based primarily on the judge's policy preferences and not the constitutional principle itself. Chief Justice Hughes' observation cannot be avoided altogether: to some extent, at least, the constitution is what the judges say it is. CHARLES EVANS HUGHES, THE SUPREME COURT OF THE UNITED STATES 120 (1928). The statement continues to be true, not because of the fact that judges construe constitutions, but because of the way in which they do.

The Court decries the prospect that our constitution should "veer in meaning each time the United States Supreme Court issue[s] a new decision", but cites with approval the change in constitutional construction occasioned by the Governor's removal of the justices of this Court following Rodriguez. The problem is the same. If a court's constitutional determinations are not in a very real sense mandated by factors distinct from the personal policy views of the justices, constitutionalism is reduced to judicial tyranny. It is no answer to say that "unsatisfactory" constructions of the constitution can be corrected simply by removing the judges. This is not the way to amend the constitution. And while it might correct the result in particular cases for a time, it leaves the process of constitutionalism itself fatally flawed. Substituting one judge for another in order to change the meaning of the constitution concedes that the constitution has no meaning apart from judges' views. The importance of individual judicial views cannot be denied, but they do not substitute for rules of law. Thus, constitutional construction must not be reduced to the issue of who is doing the construction, "we" or "they". Yet this is the result of the Court's approach, carried to its logical extreme.

The second difficulty with the Court's view that we should define fundamental constitutional rights without interference from outside the state is that it is premised on a one-dimensional view of those rights which is rarely accurate. If relator were constitutionally entitled to say whatever she pleased, this would be an easy case. But her right to speak freely is not absolute, under either the First Amendment or article I, section 8. In this case, relator's right conflicts with the district court's interest in protecting minor litigants, and the issue is whether the district court's interest warrants the restriction imposed on relator's right. This tension among competing rights and interests gives constitutional construction a multi-dimensional aspect. Thus, the Court's view that federal constitutional rights, which states cannot diminish, are a "federal safety net" is overly simplistic. Federal constitutional construction does not merely set minimum standards for protected rights which the states are free to increase; it strikes a balance among competing rights and interests that is itself of constitutional significance. While states may have more latitude in adjusting this balance than they do in reducing guaranteed protections, that latitude is not unlimited. State courts are not free from federal constitutional considerations in determining fundamental rights. The delicate balance among those rights and other interests must also be maintained.

Finally, the we-better-than-they argument evinces an inappropriate chauvinism toward the federal courts and other state courts. The concept of freedom of speech in this country did not originate in any one state, nor does any one court have a monopoly on its application. As it happens, the federal courts have been at construing the First Amendment much longer and far more often than Texas courts have had occasion to consider article I, section 8. The federal courts' experience in defining the contours of freedom of speech ought to be invaluable. To ignore it simply because we can is both imprudent and arrogant.

4

Having reviewed the Court's decision, the basis offered for it, and the defense of its new analytical method, I return to the question posed at the beginning: why should the Court go so far out of its way to invalidate the gag orders in this case on state constitutional grounds? The answer has nothing to do with the result in this case. We are unanimous in our judgment as to the outcome. The answer is not that relator's state and federal constitutional rights are different. So far as this case is concerned, they are identical. The Court's motives are ulterior. One commentator has observed that "some critics have argued that virtually all New Federalism proponents are motivated by the bare desire to achieve a liberal political agenda". Gardner, supra note 10, at 772. Whether the agenda is "liberal" or "conservative" or something else altogether makes no difference. The vice is that a non-legal influence has been brought to bear on judicial decision making. This is not "new federalism"; it is "new judicialism". The Court by its opinion today is vulnerable to this charge.

II

Relator complains that she and others have been denied access to court files concerning the pending litigation. She requests that the district court be ordered not to deny her access to these records. There are several affidavits before us, some tending to substantiate relator's position and others contradicting it. Our record contains no written order restricting access to court files, and relator does not claim that any such written order exists.

The Court rightly concludes that we cannot resolve factual disputes in a mandamus proceeding and therefore cannot grant relator's request for relief. The Court is not content to leave the matter at that, however, lest any doubt linger as to the result it intends. Thus, it adds that if the district court did restrict access to its files, it abused its discretion and violated TEX.R.CIV.P. 76a; that "there should be no impediment to viewing the court records"; that the district court should make all its records open and available to the public; and that if "[r]elator should find her access to the records in any way obstructed," she may seek additional relief. Ante, at 24. This last statement assumes that relator's access to court files has been restricted, contrary to the Court's conclusion that it cannot and has not made that determination. This rather obvious flaw aside, the Court's writing strike me as fairly heavyhanded nudging.

Rule 121(a)(2)(C), TEX.R.APP.P., requires that a certified or sworn copy of the order complained of be attached to a petition for mandamus. No such order is attached to relator's petition in this case, nor, as noted above, is one alleged to exist. If relator believed that her access to court records had been informally restricted by the district court, she should have moved for access, requested a hearing, and either obtained a ruling from the court or a record reflecting the court's refusal to rule. Without a written order or a court's refusal to issue one, this Court should neither issue mandamus nor comment on the merits of relator's complaint.

III

I agree with the Court that the district court did not abuse its discretion in removing relator as guardian ad litem of the minor plaintiffs. A guardian ad litem may be appointed for a ward only if the ward's next friend or guardian has an interest adverse to the ward's. TEX.R.CIV.P. 173. A court has no discretion to appoint a guardian ad litem for a person whose next friend or guardian has no such adverse interest, even if the court finds that appointment of a guardian ad litem would be in the person's best interest. Indeed, the next friend or guardian has a right to represent the person without the imposition of a guardian ad litem unless a conflict of interests exists. If the conflict of interest which occasioned the appointment of a guardian ad litem disappears, then it seems to me the guardian ad litem must be removed. It makes little sense that a guardian ad litem cannot be appointed without a conflict of interests between the ward and his next friend or guardian, but may continue to serve after the conflict disappears. See In re Judicial Settlement of the First Intermediate Account of Proceedings of Manufacturers Hanover Trust Co., 83 A.D.2d 808, 442 N.Y.S.2d 7 (1981) (removal of guardian ad litem permissible to save expenses when court determines that the father has no conflict with the minor); United States v. Noble, 269 F. Supp. 814, 816 (E.D.N.Y. 1967).

The best interest of a ward is sufficient cause for the trial court to replace one guardian ad litem with another. A particular guardian ad litem must be removed or replaced if the ward's best interest requires. See Barrow v. Durham, 574 S.W.2d 857, 861 (Tex.Civ.App. — Corpus Christi 1978), aff'd, 600 S.W.2d 756 (Tex. 1980) (if guardian's interests are adverse to those of child, then it is abuse of discretion for trial judge not to appoint a new guardian ad litem); Peters v. Allen, 296 S.W. 929, 932 (Tex.Civ.App. — San Antonio 1927, no writ) ("if the trial court sees that the ward's interest is not properly protected, it is the court's duty to promptly interpose in the ward's behalf to remedy the error, and, if necessary, remove the guardian ad litem and appoint another"); see also In re Estate of Lacy, 54 Cal.App.3d 172, 126 Cal.Rptr. 432, 441 (1975) (guardian ad litem who became co-beneficiary of will with minor had adverse interest and should be replaced); In re Guardianship of Lauderdale, 15 Wn. App. 321, 549 P.2d 42, 46 (1976) (guardian ad litem cannot represent two minors with conflicting interests). However, the best interests of the ward are not a necessary cause for removal of a guardian ad litem altogether.

The law presumes that it is not in the ward's best interests for a guardian ad litem to supplant an otherwise qualified parent, next friend or guardian. Even if the guardian ad litem were a more effective representative for the ward, the rights of parents, next friends and guardians cannot be set aside in this manner. Furthermore, the service of a guardian ad litem is a burden on the parties to a case. That burden is necessary when the ward's rights cannot legally be served by a parent, next friend or guardian with conflicting interests. The burden is unjustified, however, when those conflicting interests do not exist.

In the present case, the district court determined that no further conflicts of interest exist among the minor plaintiffs and their parents, next friends or guardians who would otherwise represent them in the litigation. The court explicitly stated in its order removing the guardian that because there was "no apparent conflicting or adverse interests between the Next Friends and minors . . . the appointment and retention of a Guardian Ad Litem [was] not necessary." All the parties in this latter category have settled their claims in the litigation; only the minors' claims remain. Not only are the parents and others now qualified to represent the minor children, they are entitled to do so without interference from a guardian ad litem. Therefore, I agree with the Court that the trial court correctly dismissed the guardian ad litem in this case.

* * * * * *

For these reasons, I concur only in the Court's judgment.