One of the points of disagreement between the majority opinion and the dissent was over how to interpret the Dickey-Wicker Amendment’s use of the present tense (“research in which a human embryo or embryos are destroyed…”). The dispute arises because a line of stem cells derived from a particular embryo can be kept in existence indefinitely and as a result can provide stem cells for research that is performed many years later. For example, under the Bush-administration guidelines, federal funding was available only for research projects that used stem cells that had been derived before August 9, 2001, when the Bush policy was announced. And NIH maintains a registry of stem-cell lines that qualify for use in federally-funded research. There is therefore a good chance that an applicant seeking NIH funds will use stem cells from a preexisting cell line.

The majority rejected the plaintiff’s argument that the government may not fund research using such a preexisting line of stem cells, and in doing so the  Dickey-Wicker Amendment’s use of the present tense played a big part:

 The use of the present tense in a statute strongly suggests it does not extend to past actions. The Dictionary Act provides “unless the context indicates otherwise … words used in the present tense include the future as well as the present.” 1 U.S.C. § 1. As the Supreme Court has observed, that provision implies “the present tense generally does not include the past.” Carr v. United States, 130 S. Ct. 2229, 2236 (2010). The context here does not, as our dissenting colleague would have it, indicate a different understanding. To the contrary…NIH funding decisions are forward-looking, requiring the NIH to “determine  whether what is proposed to be funded meets with its requirements.” Therefore, a grant application to support research that includes the derivation of stem cells would have to be rejected.

As a generalization about the English language, the statement that “the present tense generally does not include the past” isn’t exactly correct, but it’s close enough for government work. By which I mean that while there are in fact ways of using the present tense that “include the past,” it’s eminently reasonable to read the Dickey-Wicker Amendment as not using the present tense that way.

Most of the ways in which the use of present tense can encompass past time are irrelevant to interpreting Dickey-Wicker:

• The “historic present” which is used in narration:

A priest, a minister, and a rabbi walk into a bar…

• The “hot-news present,” which is used in newspaper headlines and the like:

Dewey Defeats Truman.

• “Past evidential use with verbs of communication”:

I hear we’re getting some new neighbors.

• Present tense used in a chronology of events:

1939    Germany invades Poland.
1941    Japan attacks Pearl Harbor.
1944    Allies launch D-Day assault.
1945    Germany and Japan surrender.

• Present tense used in captions to photos, etc.:

Barack Obama, above left, takes the oath of office.

• “Focus on present existence of works created in the past:”

Jane Austen uses a sharp satiric wit to expose follies, hypocrisies, and false truths.

• Present tense used with respect to states or habitual actions:

Paris is in France.
I take the subway to work.

• Present tense used to express “eternal truths”:

One plus one equals two.
Honesty is the best policy.

[Hat tip to CGEL for some of the descriptions and examples.]

The Dickey-Wicker Amendment’s use of the present tense obviously doesn’t fall into any of these categories.

However, there is another use of the present tense as including past time that is more pertinent—the use of the present tense in generic statements (Water is wet.) You’ll remember from the previous post about Sherley that the plaintiffs (as well as Judge Lamberth in the district court and Judge Henderson in the court of appeals) were implicitly interpreting the word research in the Dickey-Wicker Amendment as being used generically. So under that reading, the Amendment could plausibly be read as prohibiting funding of research on stem-cell lines that were already in existence when the funding was sought.

But you’ll also recall that research can also be used nongenerically, which means that the statute is ambiguous. And that, in turn, means that the courts have to defer to the government’s interpretation as long as that interpretation is reasonable. I.e., tie goes to the government.

Mark Davis, the proprietor of the Corpus of Contemporary American English (COCA) and the Corpus of Historical American English (COHA), has made another corpus available via the same interface as COCA and COHA. This one’s a little bit bigger though.

155 billion words, 62 billion of them the 1980s-2000s.

Billion with a b.  Bill-yun.

I ended my last post by noting that one of the points of disagreement between the majority and the dissent was about whether the word research could be understood to denote a “discrete project.” The majority concluded that the word as used in the Dickey-Wicker Amendment could in fact be understood in that way—an understanding under which the focus is on the specific work for which funding is sought:

NIH funding decisions are forward-looking, requiring the NIH to “determine  whether what is proposed to be funded meets with its requirements.” Therefore, a grant application to support research that includes the derivation of stem cells would have to be rejected….The definition of research is flexible enough to describe either a discrete project or an extended process, but this flexibility only reinforces our conclusion that the text is ambiguous. [paragraph break deleted]

The dissent took issue with this, describing the majority’s interpretation as a “novel” one and accusing the majority of  “strain[ing] mightily” to find an ambiguity. For the dissent, the appropriate focus was, not the specific work for which a particular application sought funding, but the field of hESC research as a whole:

My colleagues rest their…analysis on the transformation of “research” into “research project” in the Amendment’s text. In other words, it reads “research” as if it were synonymous with “research project.” But “research is the overall “systematic investigation or inquiry” in a field—here hESCs—of which each project is simply a part. Webster’s Third New International Dictionary 1813 (1993) (“project” means “a definitely formulated piece of research” (emphasis added)). [citation to majority opinion omitted]

Thus, the dissent repeatedly refers to hESC research in general:

Determining whether hESC research is “research in which a human embryo or embryos re destroyed” requires determining the meaning of “research.”

…

The first sequence of hESC research is the derivation of stem cells from the human embryo. The derivation of stem cells destroys the embryo and therefore cannot be federally funded, as the Government concedes. I believe the succeeding sequences of hESC research are likewise banned by the Amendment because, under the plain meaning of “research,” they continue the “systematic inquiry or investigation.” [citation omitted]

…

The testing and evaluation sequences of hESC research cannot be performed without first conducting the research involved in deriving hESCs from the human embryo. The derivation of hESCs is, thus, the sine qua non developmental sequence on which all subsequent sequences of hESC research rest.

In these passages, the dissent is using the phrase hESC research generically. This use is an example of one of the phenomena that come under the heading of genericity: reference to a kind of thing as opposed to a particular instance of that kind. Here’s what it says in Krifka et al., “Genericity: An Introduction”:

The underlined noun phrases NPs in (1) do not denote or designate some particular potato or group of potatoes, but rather the kind Potato (Solanum tuberosum) itself. In this usage a deneric NP is an NP that does not refer to an “ordinary” individual or object, but instead refers to a kind.

(1)     a.   The potato was first cultivated in South America.
b.   Potatoes were introduced into Ireland by the end of the 17th century.
c.   The Irish economy became dependent upon the potato.

I’ve discussed genericity previously. (A linguist walks into an app store… (part 1).) As I said there, genericity isn’t a property of a word or phrase per se, but of a particular use of a word or phrase. So a word or phrase can be used generically in one context but nongenerically in another:

Water is wet (generic)

I spilled water all over the table (nongeneric)

What this means is that the Sherley dissent’s argument that its generic interpretation represents the “plain meaning” of Dickey-Wicker—i.e., is the only way that the word research can reasonably be understood—is suspect right off the bat. And when you move beyond linguistic theory and look at evidence from actual usage, you see that the argument is not just suspect, it’s wrong.

First, research is often used nongenerically. Indeed, the plaintiffs in the Sherley case used it nongenerically in their complaint:

“[Dr. Sherley’s] research focuses on improving methods for identifying adult stem cells and producing them in large numbers for therapeutic development.”

“Dr. Sherley has received funding from NIH for research aimed at developing new methods for identification and production of human adult stem cells that have the potential for human cell therapy.”

“[Dr. Dreisher’s] research has resulted in the issuance of twenty-three patents.”

In each of these examples, research is used to refer to a particular body of adult-stem-cell research, not to the entire field of adult-stem-cell research. And it’s easy to find more examples of research being used nongenerically elsewhere:

“Haidt has conducted research in which liberals and conser­vatives were asked to project themselves into the minds of their opponents and answer questions about their moral reasoning.” [link]

“Dan Ariely of Duke University in Durham, North Carolina, accepts the [IgNobel] Medicine Prize for research showing that expensive fake medicine works better than cheap fake medicine.” [link]

(Additional real-life examples are given in my brief.)

So how does the dissent justify its (implicit) conclusion that research can only be used generically? By invoking the dictionary:

The district court correctly looked to the dictionary definition of “research” as “diligent and systematic inquiry or investigation into a subject in order to discover or revise facts, theories, applications, etc.” Research, then, comprises a systematic inquiry or investigation. And “systematic” connotes sequenced action. XVII Oxford English Dictionary 498 (2d ed. 1989) (“systematic”: “Arranged or conducted according to a system, plan, or organized method . . . .”). The first sequence of hESC research is the derivation of stem cells from the human embryo. The derivation of stem cells destroys the embryo and therefore cannot be federally funded, as the Government concedes. I believe the succeeding sequences of hESC research are likewise banned by the Amendment because, under the plain meaning of “research,” they continue the “systematic inquiry or investigation.” [some citations omitted]

There are two problems with this line of reasoning. First, the definition that the dissent quotes doesn’t support the dissent’s conclusion. There is no reason that the work done in a particular research project can’t be described as “diligent and systematic inquiry or investigation[.]” The definition is therefore ambiguous (or more precisely, underspecified) in exactly the same way as the word it defines. And second, some dictionary definitions of research—including one from the dictionary that the district court and the dissent relied on—make it unmistakably clear that the word can be used nongenerically:

“a particular instance or piece of research” (Random House)

“2a: a studious investigation or exam­ination…b (1): a particular investigation of such a character: a piece of research[.]” (Webster’s Third)

What the dissent does, then, is to overinterpret the definition it relies on and to overlook definitions that show its reading to be wrong.

Coming up next: interpreting the Dickey-Wicker Amendment’s use of the present tense.

Two things before we begin. First, a disclosure: I represent the Genetics Policy Institute as one of the amici on the government’s side in this case, and portions of these posts will be adapted from my amicus brief. Second, a point about terminology. Although the Sherley case is often referred to as dealing simply with “stem-cell research,” it actually deals with research involving human embryonic stem-cells. (hESCs). There are other types of stem cells for which research funding is not restricted. (For general background on stem cells, you can start here or here.)

Let’s start, naturally, with the statute. The Dickey-Wicker Amendment has appeared as a rider to annual appropriations bills for the Department of Health and Human Services every year since 1995. It provides:

None of the funds made available in this Act may be used for—

(1) the creation of a human embryo or embryos for research purposes; or

(2) research in which a human embryo or embryos are destroyed, discarded, or knowingly subjected to risk of injury or death greater than that allowed for research on fetuses in utero under 45 CFR 46.208(a)(2) and Section 498(b) of the Public Health Service Act…

(In granting the preliminary injunction, the district-court had relied only on subsection (1), so that was the only provision that the court of appeals discussed. Subsection (2) remains open for the district court to deal with when it gets the case back.)

A central point in both the plaintiff’s argument and the district court’s ruling was that the Dickey-Wicker Amendment not only prohibits federal funding of hESC research but that it does so unambiguously. This is important because in a challenge to an agency’s interpretation of a statute, the presence or absence of ambiguity on the relevant issue controls the extent to which the court will give the agency the benefit of the doubt. If the statute is unambiguous, the court will independently decide whether the agency’s interpretation is valid, without giving any special weight to the agency’s interpretation. But if the statute is ambiguous in a relevant respect, the court will defer to the agency as long as the agency’s interpretation is reasonable.

The plaintiff’s argument (which the district court accepted) goes essentially like this: hESC research uses human embryonic stem cells. Such cells are created (or, more specifically, derived) by extracting the stem cells from a human blastocyst (a five-day-old embryo) a process that results in the embryo’s destruction. So hESC research depends on the destruction of embryos and is therefore research in which human embryos are destroyed—even when the stem cells being used in a particular research project were derived before the project received federal funding.

The court of appeals rejected the argument that this was the only reasonable interpretation of the Dickey-Wicker Amendment’s text. This was the heart of the court’s reasoning (citations omitted; links added):

The use of the present tense in a statute strongly suggests it does not extend to past actions. [Here, the “past action" would be the derivation of the stem cells before a decision to fund research using those cells. –NG.] The Dictionary Act provides “unless the context indicates otherwise … words used in the present tense include the future as well as the present.” As the Supreme Court has observed, that provision implies “the present tense generally does not include the past.” [source] The context here does not, as our dissenting colleague would have it, indicate a different understanding. To the contrary,…NIH funding decisions are forward-looking, requiring the NIH to “determine  whether what is proposed to be funded meets with its requirements.” Therefore, a grant application to support research that includes the derivation of stem cells would have to be rejected.

The court also rejected the argument that the word research had to be interpreted broadly: “The definition of research is flexible enough to describe either a discrete project or an extended process, but this flexibility only reinforces our conclusion that the text is ambiguous.”

The dissent disagreed both with the majority’s reliance on the statute’s use of the present tense and with its conclusion that research could be understood to denote a “discrete project.” I’ll deal with each of those issues in a separate post. Stay tuned.

The decision was 2-1, and both the majority opinion and the dissent deal with a variety of language-related issues.

More later.

The issue that Leonard and Butters are opining about is whether the expression App Store is a proper name that distinctively identifies the particular location in cyberspace where one goes to get apps for one’s iPhone, as Apple contends, or whether it is a generic term for stores where one gets apps of any kind, as Microsoft argues.

I’m not going to comment here on who I think should win this fight, but I do want to make a few observation about some broader (and narrower) issues, starting with a look at how genericness (a/k/a genericity) is regarded in trademark law on the one hand and in linguistics on the other.

There is a good deal of overlap in how the concept is understood in the two fields. Speaking generally, in both fields, an expression is generic if it identifies a category of things as opposed to a particular member of the category. Now, that statement is an oversimplification in at least two ways. First, by using the term expressions, it avoids specifying the grammatical kinds of expressions that can be generic I’m going to focus here on noun phrases (NPs), because most expressions used as trademarks are NPs.  (A noun phrase is a phrase consisting of a noun and its accompanying paraphernalia, such as spaghetti, a dog, toasters, those vacuum cleaners, and that guy standing over there.) The second oversimplification is that a member of a category can itself be a category. The phrase grocery store identifies a category of stores, and the name Safeway identifies a category of grocery stores.

Despite these oversimplifications, associating “generic” with category and “nongeneric” with member of category is close enough for government blogging work. As applied to the Apple case, the category is app stores in general and the category-member is the Apple App Store.^TM

Although both the law and linguistics start out at more or less the same place, they diverge in some significant ways.

In linguistics, genericity isn’t a property of a word or phrase per se, but of a particular use of a word or phrase. That is to say, a word or phrase can be used generically in one context but nongenerically in another:

Water is wet (generic)
I spilled water all over the table (nongeneric)

So in general, at least, it wouldn’t make sense as a linguistic matter to say that it using a word or phrase generically was somehow a misuse, or conversely that word has been used generically so commonly that it could no longer be used nongenerically.

Trademark law is different. For trademark owners and their lawyers, the dichotomy isn’t so much between generic and nongeneric as between generic and trademarkable and between generic use and use as a trademark. If an expression is regarded as generic, it can’t be used as a trademark, and conversely, using a trademark generically (e.g., Coke for cola) is a misuse of the word. But if that “misuse” becomes common enough, the word becomes genericized and is no longer enforceable as a trademark. So in trademark law, unlike linguistics, genericity is a property of the word or expression itself, not of a particular use.

Or at least, that’s how it’s usually treated. However, you also see references to a trademark being used “generically,” meaning used as if it were generic (in the trademark-law sense). E.g., Here, have a Coke (said while offering a Pepsi). But even then, the usage differs from linguistics in that this use of Coke is nongeneric in the linguistic sense.

That’s all we have time for now, folks. Stay tuned, there’s more to come.

While Richard Epstein’s keynote address was interesting, it didn’t seem to me to have much to do with what the symposium was supposed to be about, and I’m at a loss to see more than the most tangential connection between what Epstein said and how his talk was described in advance. I didn’t hear any argument “that the underlying linguistic problems should drive the analysis” or that trying “to tailor rules of interpretation to institutional settings” would “detract [from] understanding how and why language works.” I don’t go to a lot of law-school symposia—I’m just a simple country lawyer—so maybe it was naive of me to expect that the program that was actually presented would reasonably approximate what was described in advance.

On the “generalist panel” Scott Soames’s presentation laid out his theory of legal interpretation but didn’t really engage with the topic until the Q&A period. Larry Solan’s presentation didn’t have that problem, but he didn’t say anything I wasn’t already familiar with from his work. (Again, was I naive to expect otherwise?) I have to confess that I haven’t watched Peter Tiersma’s presentation yet, because I’m assuming that it will cover the same ground as his paper that I linked to previously.

Things started to come alive in the Q&A period after this panel, but because of time limitations it ended too soon.

For me, the most interesting parts of the symposium are the keynote address (by Richard Epstein) and the first panel (featuring Larry Solan, Peter Tiersma, and Scott Soames.)

Epstein’s address was entitled “Plain Meaning Mostly, Right Mostly: A Modest Theory of Interpretivism.” Here’s the summary from the symposium website:

His address will discuss the attitudes toward interpretation that should be taken with constitutional, statutory and contractual materials and argue that the underlying linguistic problems should drive the analysis, and that efforts to tailor rules of interpretation to institutional settings may be useful dramatic flourishes, but in the end only detract for understanding how and why language works.

Epstein is a big name in a variety of areas, including torts and property rights, but I wasn’t previously aware of his having any particular interest in issues of interpretation. He’s a provocative thinker and I’m looking forward to seeing what he has to say.

Solan, Tiersma, and Soames are all heavily involved in law-and-language issues. Solan and Tiersma are among the small handful of law professors with PhD.s in linguistics, and Soames comes at law-and-language from philosophy. Here’s the description of their panel:

In our introductory panel, discussion will be focused on the broad question: when does plain meaning break down as a concept? Most lawyers and judges agree that the plain meaning of a text can do most, if not all of the interpretive work most of the time. Thus, another question is: why does plain meaning work most of the time? Finally, panelists will be encouraged to provide suggestions for how legal practice can be improved to avoid these interpretive dilemmas

Tiersma’s paper is the only one available online as far as I can tell. It’s entitled, “The Rule of Text: Is it Possible to Govern Using (Only) Statutes?” (h/t Legal Theory Blog.) I’ve only had a chance to skim it quickly, but it looks interesting. I hope to have more to say about it shortly.

[A] fuller treatment of the subject would have devoted more attention to whether corporations have interests sufficiently like those of natural persons to warrant interpreting “personal privacy” to encompass those corporate interests.  I think—given the overall context of the other FOIA exceptions—that the result would not have changed, but I  worry that in another case this eschewal of dynamic and purposive statutory interpretation could lead to unwarranted results.

And Dorf isn’t the only one who has complained about the decision’s textualist approach. At The Atlantic, Garrett Epps is rather harsh in his criticism:

There are, of course, different ways of deciding a statutory case like this. A judge could look to the purpose of the statute—encouraging greater transparency—and reason that extending this (at best) ambiguous exception to corporations would work against that. Or the case could turn on the legislative history of the provision–what did members of Congress and committees say when the statute was enacted? Or it could scour caselaw for analogous uses of the terms. AT&T’s brief had argued from federal caselaw and executive branch documents that, it said, supported its interpretation. The FCC had argued that extending “personal privacy” to corporations would require courts to engage in “a balancing of such newly created organizational privacy interests against the interest of the public (including some of the same shareholders and constituents) in disclosure.”

But Roberts starts and finishes in the reference section. He is to word meanings what Captain Ahab was to pale whales. Nothing in the opinion addresses any larger question of policy or jurisprudence, or even hints that he knows what the statute is for. For all the opinion shows, the Court could be a Turing machine, analyzing the structure of a cryptogram without reference to its “meaning.”

One question that may follow from these cimments is whether an intensive focus on the text (and especially an approach that draws on linguistics) is consistent with interpretive theories other than textualism.

The answer is most definitely yes.

Even if you think that the goal of statutory interpretation is to identify the legislature’s intention or the statute’s purpose, the range of possible interpretations is generally limited by the statute’s language.  (I say “generally limited” because of cases like Green v. Bock Laundry Machine Co., where even Justice Scalia endorsed a result that did violence to the text, in order to avoid an interpretation that “produces an absurd, and perhaps unconstitutional, result.”) Thus, Justice Breyer—one of the Supreme Court’s leading non-textualists—writes in his book Active Liberty that courts should look for the statute’s purpose “in difficult cases of interpretation in which language is not clear.” And it was Breyer who wrote the opnion in Muscarello v. United States, with its extensive examination of dictionary definitions, etymologies, and usage in news stories.

So since the statute’s language is always at issue, it’s always important to get the language issues right.