An empirical path to plain legal meaning

« previous post | next post »

Stephen Mouritsen, "Hard Cases and Hard Data: Assessing Corpus Linguistics as an Empirical Path to Plain Meaning", Columbia Science & Technology Law Review, 2/25/2012:

The Plain Meaning Rule is often assailed on the grounds that it is unprincipled—that it substitutes for careful analysis an interpreter’s ad hoc and impressionistic intuition about the meaning of legal texts. But what if judges and lawyers had the means to test their intuitions about plain meaning systematically? Then initial linguistic impressions about the meaning of a legal text might be viewed as hypotheses to be tested, rather than determinative criteria upon which to base important decisions.

Mouritsen's abstract continue:

There exists very little legal scholarship on corpus linguistics—the study of language function and use through large, electronic linguistic databases called corpora—and the role that corpus methods might play in legal interpretation. This omission becomes more and more striking as scholars and jurists (and even the United States Supreme Court) have found themselves persuaded by corpus-based arguments.

This Article argues that the plain or ordinary meaning of a given term in a given context is an empirical matter that may be quantified through corpus-based methods. These methods, when applied to questions of legal ambiguity, present significant advantages over existing empirical approaches to plain meaning and over the prevailing intuition-based interpretive approach of many courts. Because large, sophisticated linguistic corpora are widely available and easy to use, and because corpus methods offer a more principled and systematic alternative to the impressionistic interpretation of legal texts, corpus linguistics may one day revolutionize the process of legal interpretation.

Mouritsen's paper will enlighten legally-naive linguists and philosophers about the way that legal scholars and theorists think about the interpretation of legal language; and it will enlighten linguistically-naive legal scholars about the new opportunities for using corpus-search methods to evaluate intuitions about linguistic meaning.

A simple application of this concept was featured in Richard Posner's opinion in United States v. Costello, discussed in an earlier post.   Other LL posts on this general topic include:

"What did it mean to 'bear arms' in 1791?", 6/18/2008
"Bearing arms in LION", 6/21/2008
"The connotations of the F-word", 11/4/2008
"What does the F-word contribute?", 11/5/2008
"Q: Do corporations have "personal privacy"? A: No.", 3/1/2011
"Now on The Atlantic: The corpus in the court", 3/4/2011
"Corpus linguistics in a legal opinion", 7/20/2011
"The Corpus in the Court . . . again", 12/13/2011

Considering the large number of posts that we've devoted to other topics, and even to the category Language and the Law, this is not much — the relative frequency of posts on corpus-based legal reasoning will no doubt increase in the future.

But it's important to note that corpus-based methods, here as elsewhere, are a source of evidence for linguistic arguments, not a substitute for linguistic arguments. Without an understanding of the categories, structures, and interpretive issues involved in a given question, corpus evidence can lead you as far astray as dictionary quotations can.

This does not negate the truth and importance of Stephen Mouritsen's point about the role of usage evidence in legal reasoning about the meaning of laws, policies, contracts, and so forth. As Horace suggested more than two millennia ago, the arbiter of linguistic meaning is usage, "quem penes arbitrium est et ius et norma loquendi" ("in the power of whose judgment is the law and the standard of language"); and modern computational methods give us increasingly easy access to increasingly sophisticated surveys of increasingly large collections of increasingly varied text and speech. But I would also emphasize the importance of organizing and interpreting usage evidence in a linguistically informed way.


  1. Chris said,

    March 3, 2012 @ 11:51 am

    Agreed. However, the attempt to be "linguistically informed" has its own pitfalls. For example, I recently began reading A Penguin's Defense to the Doctrine of Equivalents: Applying Cognitive Linguistics to Patent Law by Kristen Osenga. This is a clear example of a lawyer, Osenga, trying to re-interpret the Doctrine of Equivalents (which allows broad interpretation of the wording of patent claims, a good linguistic issue) using notions borrowed from cognitive linguistics. The problem came when I read her description of cognitive linguistics (fully 1/3rd of the total article). As far as I could tell, all she did was read Lakoff's "Women, Fire, and Dangerous Things" and thought she now understood cognitive linguistics, which to her appears to mean proto-type theory and not much else. I fear she doesn't really understand contemporary cognitive linguistics (she clearly doesn't understand that proto-type theory is deeply technical and requires more than Lakoff's intro to truly understand). I appreciated her attempt and suspect her goal is a worthy one. But she falls far short of doing the leg work necessary to truly be "linguistically informed".

  2. MikeA said,

    March 3, 2012 @ 12:29 pm

    About halfway through this, I had the thought that I hope there is substantial duplication and verification of corpora, perhaps via something like LOCKSS.
    If courts start being persuaded by these arguments, there will be a real temptation to set up MiniTrue for real.

  3. PeterW said,

    March 3, 2012 @ 12:34 pm

    More than ten years ago, I did a fairly detailed (within the context) grammatical analysis of a statute in an oral argument before my state's court of appeals. Unfortunately, I've forgotten the precise details, and this was pre-Google, but the court was very interested and engaged in the process. So I'm not surprised that courts would be interested in this approach, in the right kind of case. (And one traditional legal approach to construing a statute is to look at how other courts have construed it – which is a corpora approach itself.)

  4. Rob P. said,

    March 3, 2012 @ 2:58 pm

    @Chris – The Doctrine of Equivalents seems to me to be a particularly poor place to look to linguistics for help. DOE says that where the literal meaning of the claims does not reach the accused product, we can consider whether the accused product is pretty much the same as the literally covered product. Does it perform (substantially) the same function in (substantially) the same way to achieve (substantially) the same result? That has always seemed to me to be a fundamentally technology-related question. Linguistics ought to be much more of a help one step previous, when the claim language is being interpreted to figure out what the literal scope might be. To bring it back to the original posting, when we want to know the "plain and ordinary meaning" of a patent claim term, as it would be understood by one of skill in the art, it may be very helpful to use a linguistically sound analysis of a relevant corpus to find how that term is used.

    What is often seen is reference to general usage dictionaries, or even technical dictionaries that have misleading information (for example, I've seen reference to a technical dictionary that on closer inspection was defining a term for use in technical drawing, not as it would be understood by a typical electrical engineer, or in the art of chip manufacture which was the real subject of the patent in question). These sorts of plain and ordinary meaning may well be refuted (or even supported) by appropriate corpus analysis.

  5. GeorgeW said,

    March 3, 2012 @ 3:27 pm

    Years ago I asked a lawyer friend why they used such arcane language in contracts and the like. He said that many terms have been defined in case law over the years so, everything to the contrary not withstanding, it is safer to stick with the tried and true.

  6. MikeA said,

    March 3, 2012 @ 4:51 pm

    @Rob P.: Be careful what you wish for. In my experience, the meanings attached to technical terms in patents (even my own) are pretty much orthogonal to the meanings they have to actual practitioners "reasonably skilled in the art". Patents, like contracts, are written by and for lawyers, not engineers.
    And the problem of "technical dictionaries" is becoming pretty well known on the web, where a search is as likely (or more) to return something from a "content farm" (minimum wage freelancers with word-count and buzzword quota) as something a person who might actually understand the real info might use. In other words, not all corpora are equal.

  7. UK Lawyer said,

    March 4, 2012 @ 6:41 am

    I, too, would be cautious about using the doctrine of equivalents as a case study for linguistics. Although I am not a US patent lawyer, my understanding is that it is something of a practical "fix" devised to get round an unduly narrow approach to construing US patent claims, and that there is no such doctrine in European patent laws because the basic approach to interpreting patent claims is less narrow.

    As for why there is arcane wording in contracts, that is mostly down to legal conservatism, although there are a few terms of art. The comment of GeorgeW's lawyer friend is familiar, but heard less often as the plain English campaign gathers momentum.

  8. Ray Girvan said,

    March 4, 2012 @ 1:10 pm

    @George W: safer to stick with the tried and true.

    I suspect this is the case with the UK laws on computer misuse, which lean toward defining computer-related actions in mundane concrete terms in ways that occasionally wildly stretch any 'plain meaning' interpretation. For example, the Protection of Children Act 1978 and Criminal Justice Act 1988 use "making an image" to mean downloading / viewing in a browser / opening an attachment: a definition of "make" that doesn't have much connection with the most usual sense of creating something from scratch.

  9. D.O. said,

    March 4, 2012 @ 2:52 pm

    This comment is about the title of the post. If I understand it right, plain meaning in legal context means something actually not confined to legal context, but used in ordinary speech with its ordinary (of course, contextual) meaning. Thus, plain legal meaning can not exist or at least should be something completely different. legal plain meaning, on the other hand, might mean an attempt (sucessful or not) on the part of legal community to assess the plain meaning (but maybe failing due to déformation professionnelle).

  10. Cop out II: February Link Extravaganza | polyglotpaddler said,

    March 4, 2012 @ 7:29 pm

    […] Log gives an excellent summary of articles and posts concerned with the use of corpus linguistics in legal decision-making. And if you're interested in the intersection of law and linguistics, Judge Posner is always […]

RSS feed for comments on this post