When Rules Get in the Way of Reason: One judge’s view of legislative interpretation

By Justice Paul Thissen

1119-Gavel-BirdsThe job of a judge when interpreting statutes is straightforward: “ascertain and effectuate the intention of the legislature.”1 As a former legislator, however, I am chastened to admit that the Legislature does not always make that job easy and, for a variety of reasons, too often makes that task quite difficult.2 So what is a judge to do?

This question, of course, has been hotly contested for decades. And rightly so, because the question raises fundamental constitutional and separation of powers issues. Indeed, settling on a broadly acceptable and justifiable answer to the question is critical to the legitimacy of the judiciary as an institution. In this brief article, I have no illusions of adding substance to the rich literature written by scholars more deeply steeped in the constitutional, structural, and linguistic aspects of statutory interpretation. What I can offer is a practical perspective on the task of statutory interpretation informed by my fairly unique perspective as a sitting Minnesota Supreme Court Justice, a former 16-year member of the Minnesota House of Representatives, and a former civil and criminal litigator who practiced primarily in Minnesota.

A basic challenge of statutory interpretation is making sure everyone stays in his or her lane. The job of a legislator is to sort through contested problems of public policy and pass laws to resolve, remedy, or otherwise address those problems. This is most often accomplished by imposing obligations or limitations on how the people and institutions in the state can act and organize their lives and activities. The job of a judge is to step in when a dispute arises over what those obligations or limitations mean and what impact they have on a particular party. In performing that task through statutory interpretation, however, it is essential that the judge not step out from behind the bench and into the legislative chamber to impose obligations or limitations different from those established by the Legislature.3 Stated another way, the legal contests over statutory interpretation are about how we best ensure judges stay in their proper lane.

Briefly described, the current prevailing method of statutory interpretation includes two important and interrelated concepts. First, over the past several decades, courts (and law schools that integrated this fashion of statutory interpretation in the legal curriculum) moved decisively toward a narrow focus on statutory text as the dominant (and, for some, nearly exclusive) methodology for keeping judges in their lane.4 The ability to apply rules of grammar became a more highly prized legal skill than fluency in broader policy and practical considerations of the law. Justice Antonin Scalia is perhaps the best-known advocate for this position.5 

Second, the move to a strong textual orientation in statutory interpretation emerged in the context of an increasingly rigid hierarchy of temporal rules that limits when other useful tools of statutory interpretation besides the cold text—long-established canons of construction, legislative history, clear statutory purpose, common sense—can be used. In particular, courts have erected a nearly impermeable divide between a limited and privileged class of interpretative tools that can be used to determine the meaning of “unambiguous” language and a broader set of rules that can be used if statutory language is found “ambiguous.” This hierarchy of rules cements the primacy of text as the near-exclusive basis for understanding what the Legislature meant when it enacted a law.

The central question I wrestle with in this article is this: Is the current prevailing method of statutory interpretation used in Minnesota (and United States courts generally) the best method for making sure judges stay in their lane? I will make the case that this rigid temporal hierarchy, far from enhancing a judge’s ability to do her primary job of ascertaining and effectuating the intent of the Legislature, actually impairs it by arbitrarily limiting a judge’s access to tools of statutory interpretation that may prove enlightening, even decisive, as to actual legislative intent. Moreover, the preliminary interpretive inquiry into whether statutory language is “plain” or “ambiguous” operates in a way that leaves ample space for judges to impose their preferred outcomes, but in a surreptitious, non-transparent way. Thus I worry that the strict adherence to a hierarchy of rules does not simply keep judges from entering the legislative lane; it narrows the judicial lane itself precipitously. 

I also worry about a larger issue: that strict textualism and an increasing rigidity in how tools of statutory interpretation may be applied have adversely affected how lawyers think about their job. Rules of statutory construction have become a catechism. Lawyers feel too comfortable blindly citing a rule or canon of construction without considering whether and why the rule or canon provides insight into the meaning of the particular statute being interpreted. While such an approach may make life easier for lawyers, that is bad for the practice of law. Much as I love the beauty and rationality of Euclidian geometry with its axioms and theorems, the law is not geometry and the rules and canons of construction cannot be applied that way. 

Reasoning rather than rules

So what do I offer in the alternative? In my view—and this is the central idea of this article—the job of judges is ultimately to exercise judgment.6 My intuition is that the best check on the exercise of that judgment is to demand that judges thoroughly explain their reasons for reading a statute a certain way, rather than requiring that judges follow a rigid hierarchy of rules. Crafting an explanation that seems to the parties and the public like a sensible and fair understanding of legislative intent (an explanation that may, of course, rely heavily on established interpretation rules) is a process that serves as a strong check on judicial overreach and enhances public trust in the courts. And in assessing legislative intent, judges should be able use all of the tools available to them—certainly the text of the statute, but also meaningful and relevant canons of construction, legislative history, common sense, and practical experience—as long as the judge considers and can explain why a particular tool illuminates the legislative intent behind a particular statute. 

Indeed, it is essential to the legitimacy of courts that a judge be allowed to exercise her judgment, while still being required to explain her reasons, instead of being able to simply point to a generic rule that emerged from the mists of time. How many of us have faced a situation like the day in law school I tried to go to the gym to play basketball—something I did four or five times a week—without my student ID? Even though the guy behind the desk saw me nearly every day for months, he refused to let me in. “Those are the rules,” he shrugged. No consideration of the context or purpose of the rule (to prevent people unaffiliated with the college from getting on the courts). Few phrases are more frustrating, but that is what judges do when they decide a question critically important to the parties to a case with a similarly shrugged “Those are the rules.” Judges should not rely on a hidebound repertoire of generic rules that may or may not spark any insight into the particular case. Instead we should use the best available information, whether drawn from the text or otherwise, and explain how that information provides the best insight into legislative intent in a particular case. Ultimately, this process will improve the public’s trust that the courts are truly places of fair adjudication of rights.7

Text is not everything

Of course, none of this is to say that looking first to the text is wrong. The text is often an excellent signal of legislative intent. But when statutory interpretation cases get in front of a court—and particularly an appellate court—it is likely that each party has at least an arguably coherent reason to read a statute in the way that supports its position. It is striking how often our court is faced with a statutory interpretation dispute where both parties insist that the statute is unambiguous in a way that supports their directly contradictory interpretations. That should be a clue to all of us about how plain the statute really is.8

So what gets in the way of this broader “all tools in the toolbox” approach to statutory interpretation? One major roadblock is the great ambiguity wall that judges and practitioners cite so often we can do so in our sleep: “When interpreting a statute, we give effect to the plain meaning of statutory text when it is clear and unambiguous. A statute is ambiguous only if it is susceptible to more than one reasonable interpretation, in which case we may resort to the canons of statutory construction to determine its meaning.”9 The rule creates a two-step process for statutory interpretation. First, the court must decide whether the statute is ambiguous. Only if the court decides in the affirmative can it employ most of the tools of interpretation. 

My purpose here is not to jettison the two-step process entirely. Rather, it is to call on judges to apply it more reluctantly and more humbly. My impression is that courts find statutes to be “plain” and “unambiguous” too readily. 

First, our current test for ambiguity—is there more than one reasonable interpretation of the statute?—is opaque, lacking clear and explainable boundaries. The current test creates a vast space for judges to exercise largely unrestrained judgment in deciding whether a particular interpretation is reasonable and rejecting interpretations they deem unreasonable.10 Critically, the rule provides such expansive judicial discretion even as it limits the sources of information a judge may consider for information about legislative intent. This prohibits judges from considering important signals of legislative intent, like the purpose of a statute or its legislative history, which in some cases may prove decisively illuminating. The result is a divide between pre-ambiguity and post-ambiguity reasoning, wherein judges have few tools (but broad discretion) in deciding whether an interpretation is “reasonable” for purposes of assessing ambiguity, but have access to a broad range of tools to use for guidance in the post-ambiguity realm. That makes little sense to me.

A recent case, State v. Rogers, highlighted the implications of the pre-ambiguity/post-ambiguity wall for statutory interpretation.11 Rogers required the Minnesota Supreme Court to interpret Minnesota’s burglary statute, Minn. Stat. §609.582, subd. 1(b), which provides that burglary is a first-degree offense if “the burglar possesses, when entering or at any time while in the building... any article used or fashioned in a manner to lead the victim to reasonably believe it to be a dangerous weapon.” The Court analyzed whether the victim must be physically present during the burglary for a conviction under subdivision 1(b). 

The majority, employing the Court’s traditional test for ambiguity, determined from textual clues that the victim’s physical presence was required under the statute. In particular, the Court concluded that the Legislature’s use of the phrase “the victim” rather than “a victim” meant that “the victim” must be present at the time of the burglary. The majority also concluded that the Legislature’s use of the word “used” required the victim’s presence (i.e., how could a burglar use an article in a way to lead the victim to reasonably believe it to be a dangerous weapon if the victim was not there?).12 The majority’s textual analysis was well done and reasonable. The dissent, focusing on the list of items that a burglar must “possess,” concluded from the text that there was another reasonable interpretation of the statute that did not require the victim’s presence, but the majority found that an unreasonable interpretation.13

The critical point, however, is missing from the dueling interpretations of the statutory text. More important is that the text was not the best evidence of legislative intent available to the Court. The legislative history of the statute, including the legislative debate around the amendment that inserted the phrase “any article used or fashioned in a manner to lead the victim to reasonably believe it to be a dangerous weapon” into the law, made the legislative intent clear. The amendment was adopted at a time when there was significant public concern over the use of fake weapons and the operative phrase was pulled nearly verbatim from a prior statute that used the language to refer to a fake weapon.14 In my view, this history shed much more light on the Legislature’s intent than any parsing of the statutory language ever could: The Legislature intended to make it a first-degree crime to possess a gun, a bomb, or a fake weapon. And because the victim’s presence was never required in the case of gun or bomb possession, it should also not be required for possession of a fake weapon. But because our statutory interpretation hierarchy makes resort to legislative history off-limits in assessing whether a particular reading of a statute is “reasonable” for purposes of determining ambiguity, the majority was precluded from even considering it.

The lack of clear boundaries in our current ambiguity test is exacerbated by unanswered questions about which rules can be applied pre-ambiguity and which are only relevant after a statute is found to be ambiguous.15 Take, for example, the canon that statutes in derogation of the common law are strictly construed. Minnesota Supreme Court precedent is unclear about whether the canon applies before or after an ambiguity determination is made. In some cases, the Court has applied the canon when analyzing whether a statute is ambiguous.16 In other cases, the Court has applied the rule to help analyze a statute already determined to be ambiguous.17 And although there may be reasons to justify each position, the real question is whether forcing judges to engage in such an esoteric debate gets them any closer to ferreting out the Legislature’s actual intent—or indeed gets in the way.18

In Minnesota, the current standard for determining ambiguity is also in conflict with (of all things!) state statutes. Minnesota Statutes §645.16 provides that
“[t]he object of all interpretation and construction of laws is to ascertain and effectuate the intention of the legislature” and then sets forth a pathway for doing so:

When the words of a law in their application to an existing situation are clear and free from all ambiguity, the letter of the law shall not be disregarded under the pretext of pursuing the spirit.

When the words of a law are not explicit, the intention of the legislature may be ascertained by considering, among other matters, [several factors like the purpose of the law and legislative history].19

This provision sets a much higher standard for determining that a statute is unambiguous than our current “is there more than one reasonable interpretation” test. 

According to §645.16, the words of a statute must be “free from all ambiguity” and “explicit” before it can be considered plain. And critically, the words must have such precision as applied to the existing set of facts before the Court. As I read that legislative mandate, courts should find a statute plain only if the language of the statute is so precise as applied to the particular circumstances of the case that it precludes any possibility of an alternative interpretation. Stated more simply: If there is a doubt, a judge should err on the side of ambiguity.

This restrained standard limits the discretion of judges when (under the current regime) they have access to fewer interpretive tools. Of course, using this standard allows judges to use a much wider variety of potentially useful interpretive tools in more cases. In turn, courts will reach more accurate results in ascertaining and effectuating the intent of the Legislature.20

My position also finds support in Minn. Stat. §645.08 (2018), which directs judges to construe words of a statute according to the “rules of grammar” and the “common and approved usage” of the words.21 Contrary to how the statute is generally understood, however, §645.08 conditions a judge’s resort to rules of grammar and common and approved usage with a significant caveat. Words are not to be given their common and approved usage and rules of grammar do not apply if those rules “would involve a construction inconsistent with the manifest intent of the legislature” or would result in an interpretation of the statute that is “repugnant to the context of the statute.”22 This conditional structure suggests that the Minnesota Legislature intended from the time of statehood that courts could consider more than just the plain and ordinary meaning of a word in interpreting statutes. Indeed, for several decades after statehood, Minnesota courts interpreted statutes against this general background statutory rule of construction much more broadly than we currently do.23 

Adopting an approach to statutory construction that makes more interpretive tools available more often also finds support in a recent survey of Minnesota legislators conducted by high school student Ethan Less.24 The survey was designed to learn about the tools and methods that sitting Minnesota legislators use to understand the meaning of statutes before they cast a vote.

Unsurprisingly, the Minnesota legislators reported that they relied most heavily on the text of the statute when ascertaining the meaning of bills before them in committee or on the floor. More surprising, however, was that legislators reported they relied nearly equally as much on the reports of non-partisan research and the discussion of the legislation by the chief author of the bill (and by legislative colleagues considered to be subject matter experts) as they did on the statutory text.

Legislators also reported that context matters when working to understand the legislation before them. In addition to the text, legislators responded that they found the purpose of the bill, the problem to be remedied, and the intent of the proponents of the legislation to be essential to ascertaining the meaning of a proposed law. And, notably, more than half of the legislators reported that they read the text of less than half the bills before voting on them.

According to the survey responses, the chief author of a bill is more likely not only to read the bill, but also to read the entire existing section of law when a bill only amends a portion of that section. (Legislative bills that amend only a subdivision of a larger section of existing law often only show the subdivision being amended and not the entire section.) This fact, paired with the survey results, suggests that courts should be much more open to using the chief author’s statements about a bill when ascertaining legislative intent than is the current practice. Indeed, three-quarters of the legislators surveyed said judges should rely on statements in committee or on the House and Senate floor when interpreting a statute. This compares to 100 percent who said judges should look at the words of the statute and 87 percent who said judges should consider the purpose of the statute when construing laws.

Because a judge’s job is to ascertain and effectuate legislative intent, a key message from the survey is that judges should not place too heavy a reliance on statutory text alone. Rather, judges (and lawyers trying to persuade judges) should be open to using all the tools in the interpretive toolbox to truly understand the context of a statute, at least to the extent that a particular tool makes sense in a particular case and the judge can explain why it makes sense.

“Do not use a cannon to kill a mosquito”25

The judicial duty to exercise judgment and articulate the reasons for interpretive choices extends to the specific decision to use (or not) a particular canon of construction. As Judge Posner (among others) has pointed out, “for every canon one might bring to bear on a point there is an equal and opposite canon.”26 As with the ambiguity analysis, restraint should be the watchword when applying canons of construction. Unless there are good independent reasons that a canon is useful to an interpretive project, a judge should not rely on it.

This need for rigorous judicial analysis and fleshed-out judicial reasoning rather than rote reliance on canons is borne out in other results from the legislator survey. For instance, judges and other lawyers often use the surplusage canon to determine the meaning of a statute. The canon provides that every word in a law is to be given effect and no word should be given an interpretation that causes it to duplicate another provision or to have no consequence.27 This sometimes means courts create multiple meanings to fit the rule. That’s backwards, particularly since around 90 percent of the legislators surveyed reported that statutes often are drafted with redundant terms. Three-quarters of the legislators said that in their experience, words with similar or overlapping meaning are added to a statute to make certain the meaning is clear even though the words mean pretty much the same thing. And very few legislators said they relied on the surplusage canon when trying to understand the meaning of a statute before a vote. If these results are taken seriously, it suggests that judges should have a very good reason—from context, legislative history, or otherwise—before using the surplusage canon as a decisive basis for reading a statute one way instead of another.28

Another example is the presumption of consistent usage canon: A word or phrase is presumed to bear the same meaning throughout a text.29 Although there is something intuitive about the canon, less than half the Minnesota legislators surveyed agreed that, in their experience, a word used in a statute has the same meaning throughout the statute. Further, only a quarter of the legislators reported that the meaning a word has been given in an unrelated statute is extremely or very valuable in understanding the meaning of the same word in the bill before them. Once again, these results caution judges and lawyers against relying on the consistent usage canon without thinking hard about whether and why the canon actually illuminates legislative intent when used to interpret a statute.30 Resort to context, legislative history, and common sense will be useful in making that assessment.

Legislative history is more than what legislators say

When lawyers and judges think about legislative history, we most often have in mind statements made by legislators about their intent in enacting a bill. And such statements may illuminate the meaning of a statute; they should not be ignored.

But another category of legislative history—perhaps best described as a kind of statutory archeology—is often overlooked.31 Understanding the development of a statute over time can shine a bright light on what to contemporary eyes is nothing but confusion. Don’t ignore it. 

For example, in State v. Scovel, the Minnesota Supreme Court interpreted Minnesota Sentencing Guidelines 2.B.7.a (2015), which provides that “the classification of a prior offense as a... felony is determined by current Minnesota offense definitions... and sentencing policies.”32 The central question was whether the classification of a prior offense for the purpose of calculating a defendant’s criminal history score is determined by Minnesota offense definitions at the time the current offense was committed or at the time the defendant was sentenced for the current offense. The word “current” was not defined in the Sentencing Guidelines and the multiple meanings of “current” offered in dictionaries were not helpful. What became decisive in determining the Sentencing Guidelines Commission’s intent was an analysis of the “history and evolution of Guidelines 2.B.7.a. It demonstrate[d] that the provision had its roots in Guidelines 2.B.5.b.”33 Because Guidelines 2.B.5.b made clear that “current” referred to the time when the current offense was committed (and not the time of sentencing for the current offense), the Court read Guidelines 2.B.7.a to similarly mean that offense classifications in effect at the time the current offense was committed applied when calculating a defendant’s criminal history score.

More recently, the Minnesota Supreme Court interpreted Minnesota’s research and development tax credit statute.34 The statutory formula for the credit incorporated the federal law concept of “base amount as defined in Section 41(c) of the [federal] Internal Revenue Code...”35 Unfortunately, in 2011 (the relevant tax year) I.R.C. §41(c) contained a variety of subdivisions, some of which informed the concept of “base amount” and some of which were not relevant to “base amount.” A central issue in the case was whether one of those subdivisions—I.R.C. §41(c)(2), which set a “minimum base amount” for the federal calculation—was incorporated by the Minnesota statute.

Viewed from the perspective of readers of the statute in 2011, the legislative intent in incorporating I.R.C. §41(c) was a messy thicket: Did the Legislature intend to incorporate every provision of I.R.C. §41(c) or just some of them—and if only some of them, which ones? But when one stepped back in time to 1982 (when the Minnesota R&D credit was first enacted in Minnesota) or 1991 (when the Minnesota R&D credit statute was last amended), the thicket cleared. At both points in time, I.R.C. §41(c) as incorporated into Minnesota law was a much simpler statute which included only subdivisions that informed the meaning of “base amount.”36 In particular, the “minimum base amount” limitation existed in the 1982 version of I.R.C. §41(c), removing any doubt that the Legislature intended to incorporate the “minimum base amount” limitation into Minnesota law.37

The right information leads to better decisions

In the end, we need to come back to common sense and practical experience when thinking about the job of statutory interpretation. When we make important decisions in our daily lives, we want more relevant information rather than less—and for good reason. Having the right information to thoughtfully consider typically leads to better decisions. Arbitrarily restricting the information that a decision-maker can use limits access to what may be the right information. Why should it be any different when it comes to ascertaining legislative intent?

And one last thing. As a legal community, we must do a better job of understanding how legislatures and legislators go about the practical work of enacting statutes. If judges, and the lawyers who appear before them, appreciate legislators’ methods for understanding what bills mean, we can better effectuate the intent of the Legislature. That is, after all, the job of a judge. 

Paul Thissen is an Associate Justice of the Minnesota Supreme Court. Thissen served in the Minnesota House of Representatives from 2003-2018, including as Speaker of the House, while maintaining a litigation and transactional practice in the Twin Cities. 

The author wishes to thank Ethan Less for his work on conducting the legislator survey discussed in this article. Thanks also to Anna Hall for her edits and comments on the article.


1 Minn. Stat. §645.16 (2018).

2 In my experience, there are many reasons that legislation may not be clear. Often, the lack of clarity or precision in statutory language is unintentional. Legislators are not seers and cannot anticipate every unique set of facts that may arise in the future. Ambiguity in statutory language may also emerge as established statutes are later amended. The agglomeration of new provisions on old statutes understandably produces laws that are more sedimentary than igneous. See, e.g., Minn. Stat. §609.582, interpreted in State v. Rogers, 925 N.W.2d 1 (Minn. 2019). Other factors in imprecise drafting include: the legislative process itself, which increasingly moves at a pace that does not allow staff or legislators the time to reflect and review; the increasing use of massive omnibus bills instead of discrete pieces of legislation focused on a single subject; and procedures allowing committee or floor amendments with no prior notice. A lack of transparency in the legislative process, which shuts out the public (including lobbyists, who—like them or not—have important expertise) is another factor. There are also intentional reasons for ambiguity in legislation. It is sometimes easier to assemble the votes to pass legislation when some legislators think a bill means one thing and other legislators believe a bill means something else. What judges do with these varying reasons for ambiguity is better left for future discussion. 

3 The judge’s role is certainly different when addressing the constitutionality of a statute or dealing with (vanishing) questions of common law. Those issues are left for another day. 

4 The strongest claim to legitimacy made by textualists—that it takes judicial discretion and values out of the equation—has been challenged by legal scholars and linguists alike on the ground that rules of textual interpretation and grammar can be as readily manipulated to reach the judge’s preferred outcome as any other tool of statutory interpretation. See, e.g., Richard H. Fallon, Jr., Three Symmetries between Textualist and Purposivist Theories of Statutory Interpretation—and the Irreducible Roles of Values and Judgment within Both, 99 Cornell L. Rev. 685 (Issue 4 May 2014). For a fascinating examination of how linguists view statutory interpretation, see Brian G. Slocum, ed. The Nature of Legal Interpretation: What Judges Can Learn about Legal Interpretation from Linguistics and Philosophy (University of Chicago Press 2017) (Slocum 2017). Nonetheless, the primacy of “the text” in statutory interpretation has become the water we swim in as lawyers.

5 See, e.g., Chisom v. Roemer, 501 U.S. 380, 405 (1991 (Scalia, J., dissenting). It has not always been this way. At the time of statehood, for instance, the Minnesota Supreme Court took a much broader approach to statutory interpretation. In the 1863 case of Barker v. Kelderhouse, the Court stated that “the attention of the legislature should always be followed wherever it can be discovered, although the construction seems contrary to the letter of the statute.” 8 Minn. 207, 211 (1863) (citing Grimes v. Byrne, 2 Minn. 89 (1858)); see also Rogers, 925 N.W.2d at 8, n.3 (Thissen, J., dissenting); Robert A. Katzmann, Judging Statutes 52 –53 (2014). That principle was followed and used by Minnesota courts for decades. See, e.g., Judd v. Landin, 1 N.W.2d 861 (1942) (“when [legislative intent] is ascertained the statute must be so construed as to give effect to such intention, even if it seem contrary to such rules and the strict letter of the statute”); Wegener v. Commissioner of Revenue, 505 N.W.2d 612, 613 (Minn. 1993).

6 This conception of the judicial role is captured well in David E. Pozen, Justice Stevens and the Obligations of Judgment, 44 Loy. L.A. L. Rev. 851 (2011).

7 See generally Tracey L. Meares & Tom R. Tyler, Justice Sotomayor and the Jurisprudence of Procedural Justice, 123 Yale L.J. F. 525 (2014).

8 A slight detour for a practitioner tip (with the caveat that it is coming from a single judge). I have been surprised how many times lawyers appearing before the Minnesota Supreme Court who rely on a plain language argument refuse to even contemplate the potential that the statute may be ambiguous. It is not a sign of weakness to assert that the plain language favors your client, but if the Court finds it ambiguous, your client still wins. Making an ambiguity argument in the alternative does not undermine your plain language position. Further, even if you firmly believe that the statutory language is plain, providing a judge with other information about the context, purpose, and legislative history of a statute that supports your client’s position does not hurt; it helps.

9 Gen. Mills, Inc. v. Comm’r of Revenue, 931 N.W.2d 791 (Minn. 2019). 

10 There are of course some tools that can be used “pre-ambiguity” to determine meaning. For instance, judges rely on syntactic canons like the last-antecedent canon, the series-qualifier canon, and the nearest-reasonable-referent canon when determining whether a statute is ambiguous. See, e.g., State v. Pakhnyuk, 926 N.W.2d 914 (2019); see generally, Antonin Scalia & Bryan A.
Garner, Reading Law: The Interpretation of Legal Texts 140–167 (2012) (Scalia and Garner). Judges also consult dictionaries to understand the meaning of words. See State v. Scovel, 916 N.W.2d 550, 554 (Minn. 2018). But these canons are not sacrosanct or foolproof, as our Court has properly noted. Pakhnyuk, 926 N.W.2d at 922 (stating that syntactic canons do not “trump[] the text of the statute, and… can be defeated by other indicia of meaning”); Scovel, 916 N.W.2d at 554 (stating that “it is one of the surest indexes of a mature and developed jurisprudence not to make a fortress of the dictionary”) (quoting Learned Hand in Cabell v. Markham, 148 F.2d 737, 739 (2d Cir. 1945)). Further, even resort to competing dictionary definitions is subject to manipulation. How many lawyers have scanned dictionaries to find the one definition that supports their client to the exclusion of other definitions? I know I did when I was practicing. Finally, it is worth noting that nearly 70 percent of Minnesota legislators who responded to a recent survey said they never or only rarely relied on dictionaries and other reference materials when considering the meaning of a bill before a vote. For more information on the survey, see footnote 24. 

11 925 N.W.2d 1 (Minn. 2019).

12 Id. at 4–5.

13 Id. at 7–8 (Thissen, J., dissenting).

14 Id. at 9–10 (Thissen, J., dissenting).

15 See, e.g., State v. Thonesavahn, 904 N.W.2d 432, 439 n.4 (Minn. 2017) (discussing whether the concurrence was applying the imputed-common-law-meaning canon or the common-law-abrogation canon and arguing that the former applies only after a statute has been found ambiguous).

16 See, e.g., Do v. American Family Mut. Ins. Co., 779 N.W.2d 853, 858, 859; see also id. at 860 (Anderson, P., J., concurring); Shamrock Dev., Inc. v. Smith, 754 N.W.2d 377, 382, 383 (Minn. 2008); Nelson v. Productive Alternatives, Inc., 715 N.W.2d 452, 457 (Minn. 2006). See also Staab v. Diocese of St. Cloud, 853 N.W.2d 713, 726 n.4 (Minn. 2014) (Lillehaug, J., dissenting). 

17 See, e.g., In re Stadsvold, 754 N.W.2d 323, 328–29 (Minn. 2008); Rosenberg v. Heritage Renovations, LLC, 685 N.W.2d 320, 327–28 (Minn. 2004). See Jaeger v. Palladium Holdings, LLC, 884 N.W.2d 601, 608 (Minn. 2016) (referring to the canon that statutes in derogation of the common law are to be strictly construed as an example of a post-ambiguity canon).

18 There are other reasons to question the general validity of “presumption canons” (like the canon that statutes in derogation of common law should be construed strictly) as a tool for ascertaining actual legislative intent. Indeed, as Scalia and Garner note, this particular rule is “a relic of the courts’ historical hostility to the emergence of statutory law” rather than a rule based on an understanding of legislative norms. Scalia & Garner, supra note 10, at 218. The question of how much of the common law the Legislature intended to change would be better answered by other signals of legislative intent—the text and context of the statute, the legislative history. If anything, such canons may be best justified by their role in providing broader institutional benefits to the legal system. See, e.g., Depositors Ins. Co. v. Dollansky, 919 N.W.2d 684, 696 (Minn. 2018) (Thissen, J., dissenting) (noting the benefit of alignment in common law and statutory subrogation rules). A bit more on applying canons below. 

19 Minn. Stat. §645.16 (emphasis added).

20 Supreme Court Justice Brett Kavanaugh has similarly identified the lack of clear boundaries for the threshold ambiguity inquiry as a problem. Brett M. Kavanaugh, Fixing Statutory Interpretation, 129 Harv. L. Rev. 2118 (2016). Justice Kavanaugh approaches the problem from an angle that is much more text-based than the approach articulated in this article. He proposes to do away with the threshold ambiguity inquiry altogether and suggests a two-step inquiry instead: (1) courts should “determine the best reading of the text of the statute,” and (2) “once judges have arrived at the best reading of the text, they can apply—openly and honestly—any substantive canons (such as plain statement rules or the absurdity doctrine) that may justify departure from the text.” Id. at 2135. See also Meredith A. Holland, The Ambiguous Ambiguity Inquiry: Seeking to Clarify Judicial Determinations of Clarity Versus Ambiguity in Statutory Interpretation, 93 Notre Dame L. Rev. 1371 (2018) (arguing that the Roberts Court has been more willing to consider non-textual factors in assessing whether a statute is ambiguous and applying the Kavanaugh test to Title IX cases). From my perspective, allowing courts to determine the “best” reading of a text with a limited suite of interpretive tools leaves open as much space for judicial freewheeling as the current rule. 

21 The notion that there is a single “common and approved usage” of words is itself a contestable concept. Experience tells us that, in real life, there is significant (English) linguistic diversity in our communities. We observe—and speak with and write with ourselves!—improper English grammar and usage every day. Who thinks about the last-antecedent rule of grammar in real life? I know I do not and did not when I was a legislator. And remember that legislators are enacting laws at greater speed, under greater time pressure, and with less time for review than ever before. Consequently, the notion that a “common and approved” usage exists as an objective fact is an idea that should be met with some skepticism. Karen Petroski’s article, The Strange Fate of Holmes’s Normal Speaker of English, in Slocum 2017, is an interesting path into this discussion. Moreover, in recent years, courts and lawyers have placed a handful of in-vogue grammarians on a pedestal as the final word on “proper” English grammar and usage. In so doing, courts and lawyers seal themselves off from the broader public with two consequences: (1) limiting the diversity of information that courts have access to in determining legislative intent, and (2) reducing trust and understanding of what courts do. Once again, the better approach is not just to cite the rule and the grammar or usage authority chosen by the court, but to think hard and explain why relying on a “common and approved usage” rule and, more specifically, why relying on one particular grammar or usage authority rather than another to set the standard makes sense in the particular case. 

22 Minn. Stat. §645.08(1).

23 See Barker, 8 Minn. at 211 (1863) (stating that “the attention of the legislature should always be followed wherever it can be discovered, although the construction seems contrary to the letter of the statute.”) (citing Grimes v. Byrne, 2 Minn. 89 (1858)). Language essentially identical to current Minn. Stat. §645.08 which sets forth several canons of construction was part of the Minnesota territorial statutes incorporated into Minnesota law at statehood. Minn. Gen. Stat. ch 3 §§1–2 (1858). A public meaning-originalist interpretation of Minn. Stat. §645.08 supports a broader reading of the statute than is currently in favor.

24 The survey was conducted from April to June 2019 as a senior project by Mr. Less. I served as an advisor to Mr. Less on the project. Mr. Less provided the survey to every member of the Minnesota House and Minnesota Senate. The response rate was 15%. Mr. Less also conducted follow-up narrative interviews with several legislators. The survey results are available from the author. The survey was inspired by the excellent and illuminating work of Abbe Gluck and Lisa Schultz Bressman. See Abbe R. Gluck & Lisa Schultz Bressman, Statutory Interpretation from the Inside – An Empirical Study of Congressional Drafting, Delegation, and the Canons: Part 1, 65 Stan. L. Rev. 901 (2013).

25 Attributed to Confucius.

26 Richard A. Posner, The Federal Courts: Crisis and Reform 276 (1985) (citing Karl N. Llewellyn, The Common Law Tradition: Deciding Appeals 521–535 (1960)). Judge Posner’s entire essay, Interpreting Statutes and the Constitution, which is found in the book, is well worth a read.

27 Scalia & Garner, supra note 9, at 174–180.

28 See Linda D. Jellum, Mastering Statutory Interpretation 104 (2008) (“Statutes are not always carefully drafted. Legal drafters often include redundant language on purpose to cover any unforeseen gaps or simply for no good reason at all. And legislators are not likely to waste time or energy arguing to remove redundancy when there are more important issues to address. Thus, the presumptions [underlying the surplusage canon] do not match political reality.”) Notably, other common interpretive canons like the negative-implication canon (expressio unius est exclusion alterius) find broader understanding and acceptance among legislators as they work to understand the meaning of bills before them. Over three-quarters of Minnesota legislators agreed that the concept of negative implication was useful in understanding statutes. Further, legislators were asked the following hypothetical: “If a bill refers to ‘automobiles, trucks, tractors, motorcycles, and other motor-powered vehicles’ would you assume the bill covers airplanes?” Nine out of ten legislators assumed the bill did not cover airplanes because all the vehicles listed are land-based vehicles. Similarly, Minnesota legislators broadly support the interpretive principle that doubts about the meaning of a criminal statute should be resolved in favor of the defendant.

29 Scalia & Garner, supra note 10, at 170–174.

30 Scalia and Garner agree that the consistent meaning canon “assumes a perfection of drafting that, as an empirical matter, is not often achieved” and note that the canon has some “distinguished detractors,” including Justice Joseph Story. Scalia & Garner, supra note 10, at 170.

31 See Minn. Stat. §645.16(2) (courts may consider the “circumstances under which [a statute] was enacted”).

32 916 N.W.2d at 551 (Minn. 2018).

33 Id. at 556.

34 Gen. Mills, 931 N.W.2d at 793.

35 Id. at 796 (citing Minn. Stat. §290.068, subd. 2(c)).36 Id. at 797–98. 

37 Id.