Constitutional Coup? The Case that Promulgated a New Constitution for Montana

On August 18, 1972, the Montana Supreme Court, in a 3-2 decision, issued its judgment in State of Montana ex rel. Cashmore v. Anderson.

500 P.2d 921 (Mont. 1972), cert. denied, 410 U.S. 931 (1973).

The Cashmore case was distinctive for other reasons as well. Rather than allowing the case to work its way up the judicial hierarchy, the Montana Supreme Court granted a request that the court dispose of it immediately by exercising original jurisdiction. Even when a dispute over determinative facts arose shortly before the scheduled hearing the court retained the case rather than remit it to a fact-finder. Then without providing the losing side sufficient time to respond to the new factual issues, the court held a hearing limited to legal issues and soon thereafter issued its decision.

Before the case arose, there was an almost-universal understanding of the specific voter majority required for approval of a new Montana constitution. Cashmore not only abandoned that understanding for a different one, but did so after the referendum already had been held.

For a case of such consequence the majority and dissenting opinions were oddly drafted. They were indifferently researched, curiously disorganized, internally inconsistent, and occasionally incoherent. The dissent showed signs of being patched together at different times and under different circumstances. Some Montanans in a position to know believe that one justice on the five-man court changed his mind after initial drafts were prepared, thereby forcing hasty redrafting. Some claim the vote switch was the product of political pressure.

Such a case cries out for scholarly review. But there has been almost none in the 46 years since the constitution was proclaimed. Montana’s principal organ of legal analysis, the Montana Law Review, has published almost nothing on the subject.

The Review’s sole treatment has been a three page discussion in an article on another topic, not by a legal scholar but by a political scientist. See Ellis Waldron, The Role of the Montana Supreme Court in Constitutional Revision, 35 Mont. L. Rev. 227, 259-61 (1974).

Waldron’s attitude toward the 1972 constitution was not one of unbiased scholarship. He was a zealous advocate, having served as a consultant to the Legislative Council when it developed its report assailing the 1889 constitution, Occasional Paper No. 6, supra note 1, at ix, was a member of the Constitutional Convention Commission, Occasional Paper No. 1, supra note 1, at ii, and wrote its report on legislative reapportionment. Constitutional Convention Commission, Legislative Reapportionment, Memorandum # 10 (1972).

I had published several times in the Montana Law Review when in 2007 I offered to produce an examination of Cashmore. The Review’s editors declined the offer as too controversial.

The silence on Cashmore has been accompanied by much celebration of the 1972 constitution itself.

E.g. Fresh Chance Gulch, Time Magazine, Apr. 10, 1972 (referring to the 1889 constitution as “creaky” and referring to the new one as a “model”); Kristen Inbody, MT Constitution Lets the ‘Sunshine in,” Great Falls Tribune, Oct. 31, 2014 (celebratory “news” story). Similar favorable treatment pervades the only book on the constitution. See generally Elison & Snyder, supra note 1.

Praise for the constitution and its framers frequently approaches hagiography, e.g., James C. Nelson, Keeping Faith With the Vision: Interpreting a Constitution for This and Future Generations, 71 Mont. L. Rev. 299 (2010) (former state supreme court justice) (“It is, in my view, the most progressive, people-friendly, and pro-civil-rights organic document of any state constitution”; id. at 301; “I firmly believe that Montana’s Constitution is the finest, most progressive state constitution in the country,” id. at 322).

Before the election, the convention leaders had shared the common understanding that “a majority of the electors voting at the election” meant a majority of all voters participating.

Supra notes 150-55 and accompanying text.

Overruling the scruples of Frank Murray, the Democratic Secretary of State,

Atlas, supra note 1, at 259.

U.S. District Judge Charles C. Lovell, who as a young lawyer argued Cashmore for the state attorney general’s office, says he was never quite sure of Anderson’s motives in signing—whether he believed the constitutional majority standard didn’t mean what everyone said it did, or whether he was putting on a political show.

Oral Conversation with U.S. District Judge Charles C. Lovell, Helena, Aug. 24, 2018.

Contemporaneously with the signing, lawyers for William C. Cashmore, a Helena physician, and Stanley C. Burger, executive director of the Farm Bureau,

Dr. Cashmore had served as a Republican in the state senate from 1961 to 1963 and in the house in 1969 and again in 1971. Ellis Waldron, Montana Legislators, 1864-1979: Profiles and Biographical Directory (Bureau of Gov’t Research, University of Montana, 1980). Dr. Cashmore ran unsuccessfully for constitutional convention delegate in 1971. Atlas, supra note 1, at 257. Burger was a long-time conservative activist who founded the Montana Farm Bureau as a more conservative alternative to the liberal Farmers Union. Telephone Conversation with Tom Rolfe of Helena, Montana, Aug. 6, 2018.

On June 28, the governor filed his answer and a supporting brief.

Answer and brief available at https://i2i.org/court-papers-in-montana-ex-rel-cashmore-v-anderson/

As in the referendum campaign, the constitution’s opponents found themselves outgunned. Intervening to support the governor and the attorney general were the City of Billings, Montana’s largest municipality; Robert L. Kelleher, a lawyer who had served as a convention delegate; and a group of former delegates led by Leo Graybill, Jr., the convention president. Submitting amicus curiae briefs on the same side were five organizations, including Common Cause and the League of Women Voters.

Register of Action in the Supreme Court of the State of Montana, Case No. 12309, available athttps://i2i.org/wp-content/uploads/Cashmore-Register-of-Action.pdf;see also Associated Press, 17 Attorneys to Air Views on Constitution, Great Falls Tribune, Jul. 7, 1972, p.17.

All briefs filed in the case are available at https://i2i.org/court-papers-in-montana-ex-rel-cashmore-v-anderson/

The relators soon faced a more serious disadvantage. The court had assumed original jurisdiction on the premise, accepted by all, that there was no factual dispute. The relators, their supporting intervenors, and their allied amici had prepared their briefs on that supposition. Then, on July 12—a scant five days before oral argument and more than three weeks after receiving permission to intervene—the attorney general filed his formal answer.

Answer of Attorney General, available athttps://i2i.org/court-papers-in-montana-ex-rel-cashmore-v-anderson/

The attorney general’s answer, backed by a brief filed two days later,

Brief of Attorney General, available athttps://i2i.org/court-papers-in-montana-ex-rel-cashmore-v-anderson/

Apparently several intervenors and amici allied with the governor knew in advance that the attorney general had this surprise planned, because within two days the Graybill intervenors, the League of Women Voters, and Common Cause all had filed briefs focusing on the new factual dispute.

Briefs available athttps://i2i.org/court-papers-in-montana-ex-rel-cashmore-v-anderson/

Under these circumstances, the court could have pursued any of three defensible courses. The best would have been to remit the case to a trial judge for a hearing on the factual question—and, preferably, for development of the legal issues as well. The second best would have been to employ a special master to resolve the factual question. A barely-defensible option would have been to postpone oral argument and afford the relators time to investigate and respond. But the court adopted none of these courses. Instead, it retained original jurisdiction and proceeded with oral arguments as scheduled.

Those arguments were held on July 17, with numerous convention delegates peering from the courtroom galleries.

J.D. Holmes, 237,000 Key Figure in Constitution’s Fate, Great Falls Tribune, Jul. 18, 1972, p. 1; see also Frank Adams, Constitution Backers Lean to Optimism, Great Falls Tribune, Jul. 18, 1972, at 1 (reporting that at least 30 delegates were in the galleries).

On the face of it, those predictions seem to have been confirmed.

Frank Adams (column), Great Falls Tribune, Aug. 27, 1972, at 25. But see infra notes 303-08 and accompanying text (proposing the alternative hypothesis that Justice Haswell was the swing vote).

In view of the importance of the case and its departure from previous authority, it would be gratifying to report that Justice Haswell’s opinion was rigorously researched, carefully written, and powerfully argued. Unfortunately, such a report cannot be made. Indeed, the opinion’s organizational defects are such that reorganization is necessary before analysis can be attempted. When reorganized, Justice Haswell’s opinion coalesces into six fundamental propositions:

We consider each point, in turn.

On this subject, Justice Haswell wrote for the court:

In fact, there were not “two distinct and opposing lines of authority.” Those cases equating “a majority of electors voting at the election” with “a majority voting on the question” arose from single issue special elections. In other words, they differed only in the scope of the election, not in the meaning of “a majority of electors voting.”

Supra note 59 and accompanying text.

This passage seems to have served the rhetorical purposes of dismissing all authority but two Montana cases on which the court’s majority wished to rely.

Having disposed of other authority, Justice Haswell opined that “we must consider the policy and philosophy of government contained in our Constitution as enunciated in numerous [Montana] cases … . ”

500 P.2d at 928.

Tinkel involved a one-issue special election, so the number of votes in the election was identical to the number of votes on the issue.

68 P. at 860.

As noted earlier,

Supra note 244 and accompanying text.

Morse v. Granite County

119 P. 286 (Mont. 1911).

Justice Haswell must have understood that neither Tinkel nor Morse dictated the answer in Cashmore.

Certainly this was pointed out in several briefs. E. g., Brief of Petitioner, The State of Montana ex rel. v. Burger (Cashmore), available athttps://i2i.org/court-papers-in-montana-ex-rel-cashmore-v-anderson/ at 15-17 (discussing Tinkel and Morse); Amicus Curiae Brief of Gerald J. Neely, available at id. at 21 (discussing Tinkel).

There is a presumption that when a legal document employs different phrases the phrases carry different meanings.

E.g. Henson v. Santander Consumer USA, Inc., 137 S.Ct. 1718, 1723 (2017).

Yet this passage adds text to the 1889 constitution that was not present. It states, “The referendum to the voters on the calling of a constitutional convention is normally held at a general election as was done here,” but nothing in the constitution so required. It required only that the legislature “submit [the proposal for a convention] to the electors of the state.”

Mont. Const. of 1889, art. xix, § 8.

Thus, Justice Haswell’s opinion created distinctions between elections that did not exist. And without them, no reason remained for disregarding the presumption that different language signifies different meaning.

“We are mindful of the principle,” Justice Haswell wrote,

Of course the differences between the parties arose not from differences about majority rule but about the group from which a majority is determined. The dispute was over decisional denominators rather than numerators. More importantly, perhaps, Justice Haswell cited no authority for the proposition that majority rule is a matter of natural right. On the contrary, one reason super-majority requirements appear in constitutions is to better protect the “natural rights” of individuals and minorities.

E.g., John O. McGinnis & Michael B. Rappaport, Our Supermajoritarian Constitution, 80 Tex. L. Rev. 703, 722 (2002) (“Both the large republic and supermajority rule founded government on popular consent yet reduced the power of particular factions to oppress the rights of minority opponents.”).

Near the end of his majority opinion, Justice Haswell alluded to the factual issue raised by the attorney general: The secretary of state reported 237,600 electors as voting, but that figure may have included all those receiving ballots rather those who cast them properly.

Cashmore, 500 P.2d at 930.

Certainly the court should have yielded to the facts, but it neither ordered a recanvassing nor appointed a fact-finder to determine what those facts were. Instead, the majority opinion insisted that

Accordingly, the majority added together the number of votes on the most-voted for issue in each county—that is, the gambling question in 18 counties and the constitution in the remaining 38. This yielded a total of 230,588 voters. The constitution, wrote Justice Haswell, was ratified under the traditional rule because the affirmative vote of 116,415 represented a majority of 230,588.

Id.

Unfortunately, these calculations demonstrated a lack of numerical understanding. In absence of specific legal authorization, one cannot employ the most-voted-on question as a proxy for total votes cast because some electors opt to vote only on issues other than the most-voted-on question. Consider the hypothetical five-voter election posited above:

Supra note 48 and accompanying text.

The most voted-on candidate or issue is Proposition A—four votes. But that was not the number of electors who voted (five), because Elector 3 voted for the candidates and on Proposition B, but not on Proposition A.

Nor was this a merely theoretical concern in the actual referendum:

In other words, the court’s estimate omitted every elector in 36 counties who cast a ballot but decided not to vote on the constitution and every elector in the other eighteen counties who cast a ballot but decided not to vote on gambling. The number omitted may have been very significant—but even an undercount of less than one percent would have raised the denominator sufficiently to depress the constitution’s percentage below the necessary majority.

If the number of “electors who vote[d] at the election” was higher than 232,831, the constitution would not have had the necessary majority. The 2243 figure is 232,831 minus 230,588, the number of voters conceded by the court’s opinion.

Chief Justice James T. Harrison’s dissenting opinion

Cashmore, 500 P.2d at 930-45.

As finally issued, the dissenting opinion began with a preface of about 660 words. This preface was haunted by a spirit of exasperation. It recited earlier proceedings and complained that the majority was not acting in a consistent manner. It specifically criticized Justice Haswell for contradicting his own statement in an earlier case.

Id., 500 P.2d at 931.

After the preface came the core exegesis. It was composed in the indicative mood.

The indicative mood is a grammatical term used to indicate the meaning of verbs. As illustrated by the text infra, it is to be distinguished from the subjunctive mood. The indicative mood is used, inter alia, for statements of fact, e.g., “I see a giraffe.” The subjunctive is used, inter alia, for statements contrary to fact, e.g., “I see only an elephant; if I saw a giraffe I would tell you.”

From the overall structure it appears that the core exegesis was to be the majority opinion, but when it became clear the writer or writers did not command the majority, he (or they) interlineated the 150-word passage, added the frustrated preface, and appended the testy conclusion.

Supporting the hypothesis of piecework composition are some other oddities. One passage was incoherent and another bore no relation to the remainder of the text. The incoherent passage appeared in the portion of the opinion that agreed with the holding in Tinkel:

Perhaps the writer intended to say he agreed with Tinkel insofar as it limited the decisional denominator to those who voted in the special election, but that Tinkel was inapplicable because the proposed constitution may have received less than a majority of those who voted in the special election.

The passage without relation to anything else in the opinion was as follows:

This passage appears to have been dropped into the opinion by mistake. There is no other reference to “positive assent” in either the majority or dissenting opinions. The passage derives from the part of Neely’s brief that explained the “majority of electors voting at the election” requirement as one that ensured that ratifiers approve by positive assent rather than by silence.

Amicus Brief of Gerald J. Neely, supra, at 3 & 40.

The coherent portions of the dissent’s core exegesis cited and reproduced extracts from constitutional provisions and from eight decided cases defining “a majority of electors voting at the election.” It then distinguished Tinkel and Morse, and finally discussed the factual question of how many voted in the special election.

Cashmore, 500 P. at 939-43.

Despite its length, the dissent suffered from several lost opportunities. First, its author(s) should have contended that original jurisdiction had been improvidently granted or, once granted, should have been revoked after the factual issue surfaced. The factual issue—and, indeed, the complex and important legal issues—justified careful consideration at the district court level, or at least by a special master.

Second, the dissent should have noted the unfairness of the proceedings: Five days before the hearing—after all parties had agreed that there was no factual dispute and after the briefs of the relators and their allies had been prepared—the attorney general and his allies produced a factual dispute. Under the circumstances, the court should have postponed the hearing, asked the relators for briefs on the factual issue, or otherwise permitted an opportunity for response.

Third, the dissent strung together a list of relevant cases, but failed to draw two necessary conclusions: One was that the phrase “a majority of electors voting at the election” had a clear, accepted meaning, not a disputed or debatable one. The other was that this was also the meaning when the voters ratified the 1889 constitution. It was this understanding that should have governed the case, not the “philosophy” of the Tinkel dicta issued thirteen years later.

Fourth, the dissent failed to show how that meaning, and the public message that an abstention meant “no,” invited those who opposed the constitution to abstain.

Fifth, the dissent failed to mention how the manipulated structure of the paper ballot could have affected the election results. For example, contrary to the standard practice (followed for the other three propositions) that a “yes” vote is a vote to alter the status quo, the ballot provided that a “yes” vote continued the death penalty. It also advised voters that “If the proposed constitution fails to receive a majority of the votes cast, alternative issues also fail.”

Id., 500 P.2d at 923 (reproducing ballot form).

Finally, the dissent failed to challenge the majority’s erroneous claim that the number of electors casting ballots on the most voted-upon issue was equivalent to the total number of electors voting.

On September 5, Burger filed a petition for rehearing. The petition included extensive argument. Much of it represented a futile effort to persuade the court to reconsider its legal conclusions, but it also included the first written rebuttal of the attorney general’s claim of factual dispute. The petition pointed out that county election officials copied the voter numbers they sent to the secretary of state from their “poll books,” and that under state law a poll book recorded only those who actually cast valid ballots, not everyone who was issued a ballot. Attached to the petition were affidavits from two county clerks affirming that the numbers they transmitted represented only those who had properly voted.

Petition for Rehearing, available athttps://i2i.org/wp-content/uploads/1972-0905-Rehg-Petition-Burger.pdf

The Burger petition also featured elaborate statistical examples showing that the number of votes on the most-voted-upon issue was not the same as the total number of electors participating in the election.

The following day, Dr. Cashmore also filed a petition for rehearing. Cashmore’s petition noted that the constitutional referendum was part of a multi-issue special election, and discussed cases arising in such elections. It also urged a “recount” of the vote.

Petition for Rehearing of the Relator, William F. Cashmore, M.D., available athttps://i2i.org/wp-content/uploads/1972-0906-Rehg-Petition-Cashmore.pdf

The attorney general’s response accused the relators of trying to re-litigate issues the court already had decided.

Objections to Petitions for Rehearing, available athttps://i2i.org/court-papers-in-montana-ex-rel-cashmore-v-anderson/; Memorandum in Support of Objections to Petitions, available at id.

The denial was, in retrospect, probably inevitable. The petitions contained little more on the legal issues than that already offered by the parties, intervenors, or amici in the earlier proceedings. The Burger petition, it is true, demonstrated clearly the error in equating the vote on the most-voted-on issues with “electors voting at the election.” Additionally, it cast doubt on the conclusion that there were fewer than 237,600 actual voters. But the statistical portion of the court’s opinion had been dicta anyway.

After the Montana Supreme Court issued its decision on rehearing, Dr. Cashmore surrendered, as he earlier had announced he would.

Tribune Capitol Bureau, Document Ruling Seen Acceptable, Great Falls Tribune, Jun. 24, 1972, at 7 (“But, says Cashmore, ‘we should settle things in Montana without going to Washington. Certainly a matter of this kind should be settled here.’”); Associated Press, Challenger of Constitution Accepts Court Ruling Despite Disappointment, Great Falls Tribune, Aug. 19, 1972, pat 3.

However the district court found no Fourteenth Amendment violation:

The court added that the newspaper supplement was an “unofficial” document and that if “any electors were in fact misled, they were simply mistaken as to the effect of their abstention from voting and not deprived of any right or opportunity to vote … . ”

Id. at 512

Although most election errors do not constitute Fourteenth Amendment violations, this decision is somewhat disquieting. Surely the Montana public was entitled to assume the constitutional language would be interpreted as represented by all concerned, with abstentions being counted as “no” votes. Perhaps Montana officials can be accused of changing settled election rules after the election was over—which surely is a Fourteenth Amendment violation.

Cf. Bush v. Gore, 531 U.S. 98 (2000) (voting, once recognized by the state, is a fundamental right, and voters must be treated equally; shifting rules after the election is inconsistent with the Equal Protection Clause); Griffin v. Burns, 570 F.2d 1065 (1^st Cir. 1978) (finding due process violation where ballots cast in accordance with existing practice were invalidated after the election by retroactive application of new rule).

When the court granted original jurisdiction it did so upon the representation by all parties that there were no unresolved issues of fact. Once it became clear this was not true, a prudent tribunal would have remitted the case to a trial judge or at least to a special master. If the trier of fact found that a critical number of ballots issued were not validly cast, the constitution would have been ratified under the traditional rule, and there would have been no need to spend court time on exhaustive treatment of the law. Even in the absence of a factual dispute, the case could have benefited from lower court review because of the extensive amount of case law interpreting the phrase “a majority of the electors voting at the election.”

Why did the court retain original jurisdiction in such circumstances? Cashmore is not the only case in which the Montana Supreme Court’s exercise of original jurisdiction amounted to judicial malpractice,

See Marshall v. State of Montana, 975 P.2d 325 (Mont. 1999), in which the court took original jurisdiction of challenge to voter-passed constitutional initiative without inquiring into the standing of the plaintiffs, and invalidated the initiative by altering a ballot rule after the election was held.

The answer to the latter question may be simply because Charles C. Lovell from the attorney general’s office, funneling his appeal through the words of Montana’s longest-serving chief justice, out-argued the constitution’s opponents. Otherwise, the two questions may have some common answers. First, there is a substantial body of research showing that the decisions of judges, in particular elected judges, are influenced by personal incentives and judicial self-interest.

Joanna Shepherd, Measuring Maximizing Judges: Empirical Legal Studies, Public Choice Theory, and Judicial Behavior, 2011 Ill. L. Rev. 1753 (2011) (summarizing recent findings, including the influence of such factors as length of tenure, judicial elections, whether the elections are partisan or non-partisan, fear of reversal, and campaign contributions).

It is not disrespectful to the Montana Supreme Court to observe that the justices had powerful incentives to short-circuit the judicial process and uphold the 1972 constitution. The new constitution’s omission of fiscal limitations could be expected to increase funding for the judiciary. The greater scope for legislation and some of the new constitution’s open-ended language promised more judicial business and more scope for judicial efforts to “do good.” The new constitution also extended the justices’ terms of office from six years to eight.

Compare Mont. Const. of 1889, art. vii, § 7 (six years) with Mont. Const. art. vii, § 7 (eight years).

Additional influences on the justices may have arisen from their daily associations: They were, after all, public employees and human beings. They were in the hub of a county that awarded the new constitution the second highest percentage of any county in the state.

Atlas, supra note 1, at 262 (showing that Lewis and Clark County awarded the new constitution over 59 percent of the vote, second only to Cascade County (Great Falls)).

The justices may have been subtly affected also by the foreseeable consequences of alternative outcomes. It seems all but certain that the constitution’s advocates would have “punished” an adverse decision, perhaps with continued litigation and perhaps through mass media favorable to the new constitution. A decision against the constitution, on the other hand, entailed fewer costs. The constitution’s opponents had demonstrated their media ineptitude during the ratification campaign, and Dr. Cashmore had made a tactically unwise public statement ruling out in advance any federal court proceedings.

Supra note 284 and accompanying text.

In such circumstances it is not remarkable that three justices voted to short-circuit the process, disregard precedent, and rule the constitution ratified. It is perhaps more remarkable that two justices did not.

Why, then, having decided to retain original jurisdiction and abandon the traditional rule, did they not take more time and care in organizing and composing their opinions?”

Part of the answer to this question may lie in the proceedings within the court’s chambers. Some insiders claim the initial vote among the justices was 2-3 against the constitution, that each side prepared an opinion on that basis, that one justice switched sides, and that both opinions had to be re-written.

Telephone Conversation with Robert Campbell, Aug. 13, 2018.

If the initial vote was 2-3, then who switched? The person usually identified is the affable John C. Harrison,

Telephone Conversation with Robert Campbell, Jul. 30, 2018; Telephone Conversation with Robert Campbell, Aug. 13, 2018. Campbell remains active in post-convention memorial events.

But several reports nevertheless identify John C. Harrison as the justice who changed his vote. According to a prominent constitutional convention delegate, someone leaked the story of the 2-3 preliminary tally (with Harrison voting against) to U.S. Senator Lee Metcalf, a strong advocate of the new constitution. Accordingly, Metcalf confronted Harrison in Helena and threatened to run him out of the state Democratic Party if he voted to (in his words) “kill the constitution.”

Telephone Conversation with Robert Campbell, Jul. 30, 2018; Telephone Conversation with Robert Campbell, Aug. 13, 2018.

If tampering did occur, then it further explains why the justices did not draft their opinions with more care. There is not much incentive one to take pride in one’s written product when that product is not really your own.

There is, of course, no chance the Montana Supreme Court will reverse the result in Cashmore. In fact, the court is highly protective of the 1972 constitution.

E.g., Marshall v. State of Montana, 975 P.2d 325 (Mont. 1999); Montana Ass’n of Counties v. Montana, 404 P.3d 733 (Mont. 2017) (together rendering significant amendments almost impossible).

One such reform would be to adopt rules that prevent state supreme courts from assuming original jurisdiction except in the most dire emergencies. Even in emergencies a case with unresolved factual issues is never appropriate for original jurisdiction without a mechanism for reliable fact-finding. After all, we have lower courts for a reason: They resolve factual issues in hearings specially designed to do so, and they clarify legal issues for higher tribunals. There was no real emergency in the Cashmore case, no reason it should not have been examined first by a trial court, and every reason to believe it should have been.

When appeals courts do exercise original jurisdiction, they should apply the usual standards for late-breaking evidence—that is, exclude it or provide the opposing party with a fair opportunity to respond. This would seem to be a basic requirement of due process.

More fundamentally, legal reformers should examine seriously the effects of incentives on judges and how to address them so as to better preserve judicial impartiality. For example, jurists who live in a capital city are as likely as anyone else to be caught up in the thinking that prevails in that capital city. When justices of a state’s highest court are reviewing the validity of a measure from which they, and the institution they work for, will benefit or suffer, they face a conflict of interest.

This Article does not focus on judicial reform, and I do not, therefore, offer comprehensive solutions. At the least, however, we might abandon the dogma that the state’s highest court must be located in the same place as the governor, legislature, and central bureaucracy. The Montana Supreme Court might be better located in Billings or Great Falls, or perhaps rotate between the two, rather than in the small-town political hot-house that is Helena.

Existing mechanisms for temporarily replacing judges who face conflicts of interest can be more broadly applied. Replacement could come from the ranks of district judges. Alternatively, or in addition, Montana could team with other low population states to form an interstate pool consisting of judges of unusually high quality. Montana could draw from the pool for impartial judges from other states in cases in which state supreme court justices face conflicts and the matter for determination does not require detailed knowledge of Montana law.

I am indebted to Andrew P. Morriss, Professor of Law and Dean of the School of Innovation at Texas A&M University, for this suggestion.

More important than any specific suggestion, however, is the lesson that the rule of law is a fragile thing, and easily shattered when those in power find the reasons for shattering it sufficiently appealing.