That, Which

The subject of vast confusion.
Christopher Lasch

On at least one point of grammar, American English turns out to be superior to its competitors. The relative pronouns "that" and "which", when they turn up in what the Chicago Manual of Style calls "polished American prose", have clear and well defined uses. "In British English," by contrast, "writers and editors seldom observe the distinction between these two words." (5.202, p. 230)

This is confirmed by Pam Peters' Cambridge Guide to English Usage, where she notes that "which often provides an alternative to that in reference to things," and that in most cases "the choice is purely stylistic". She does point out, however, that even in British English, "the choice may be influenced by the nature of the clause [the pronoun] introduces -- whether it is 'restrictive' or 'non-restrictive.'" (Pp. 576-7) It is the difference between the restrictive and non-restrictive use of these pronouns that I want to emphasize here.

Consider the following example.

(1) In the summer of 1993, Dale Jones and Bill Ewing met for the last time. The meeting, which would prove to be a historic one for Altern Technologies, was held at the Emory Hotel on the banks of the Mississippi.

Here the clause that begins with "which" does not restrict the meaning of the words "the meeting", which has been adequately determined by the previous sentence. Note the commas that set it off from "The meeting" and "was held", which indicate that the sentence could survive the removal of the clause in question. We would then have, "The meeting was held at the Emory Hotel on the banks of the Mississippi." That is, the clause is effectively parenthetical.

Here is an alternative formulation.

(2) The meeting that changed the course of Altern Technologies forever was also the last time Jones and Ewing spoke face to face. It was held at the Emory Hotel as a heavy July rain fell into the Mississippi.

If we try to remove the qualifying clause introduced with "that" we see how much work it is doing. "The meeting was the last time Jones and Ewing spoke face to face," is all we'd be left with. It is unclear what meeting is being talked about in this sentence, though the definite article ("the") tells us that that it is not just any meeting.

Ironically, we use the word "that" to specify which meeting we are talking about, while the word "which" is best used when we are simply providing extra details about a meeting that has already been clearly identified.

So the rule can be summarised as follows. Use "that" when you want to restrict the meaning of the noun it follows and use "which" (preceded by a comma) when you are providing additional information about the noun which does not restrict its meaning. Thus,

(3) The meeting that I told you to attend has been cancelled.

But not,

(4) The meeting which I told you to attend has been cancelled.

Note, however, that this is what British writers and editors care less about. You should certainly not write,

(5) The meeting, which I told you to attend, has been cancelled.

No one recommends this punctuation if what you mean is (3). But you might, of course, find yourself needing so say something like the following.

(6) The meeting, which I told you to attend, was very important. Why did you disobey me?

But note that the words "which I told you to attend" are not intended to restrict which meeting is being talked about. That should already be clear by the time this point is being made.

Note that (6) does not mean the same thing as this:

(7) The meeting that I told you to attend was very important.

The difference here may seem needlessly subtle, but it is the purpose of a grammatical rule to allow you to be as subtle as you need to be for given purposes. (7) means "Remember the meeting I told you attend? It was very important." The one before that, however, means something different: "The meeting was very important and I told you to attend it." While the fact that I told you to attend the meeting restricts the meaning of the words "the meeting" in (7), it states an independent fact (6).

Here is one last example of the correct use of "which".

(8) All the top managers at Altern were immediately summoned. The meeting, which was to be held the following Thursday, was soon cancelled, however, because Dale Jones was unable to attend.

Finally, note that American usage does not contradict British usage on this point. American usage is always correct under British rules. British usage is more liberal, allowing "which" to replace "that" in (3), giving us (4), while American usage does not. While (4) sounds odd to a "polished" American reader, (3) only sounds less pretentious to a British one. I recommend learning the American rule because it amounts to easily distinguished rules for "that" and "which" when used as relative pronouns. It is less likely to lead to confusions in other cases.

For Peters rightly calls "that" the "workhorse of the English language." It can, for example, also be used as a demonstrative, e.g., "I went to the party. That was a mistake." (Note the difference punctuation makes: "I went to the party that was a mistake.") And as an adverb: "I didn't think there'd be that many people there." But more on that later.

Abstract Expressionism: the art of the nutshell

I could be bounded in a nutshell and count myself a king of infinite space--were it not that I have bad dreams.

Hamlet

Writing an abstract can be a nightmare. Many academics complain about the absurdity either of summarising a paper they have not yet written or of summarising a paper of 5000 or 8000 words in 300 or 500. Either way, they say, it is an impossible task to 'get the idea across' without employing the expressive resources of a full paper.

I hear this often, so I want here to defend the abstract as an academic genre; indeed, I want to sing its praises as an essential tool, not just of the communication of research results, but as an integrated part of the research process. At bottom, contempt for the abstract reduces to contempt for the act of writing itself; it is a denigration of the power of words as such.

Stéphane Mallarmé famously noted that poems are not made of ideas but of words*. He thereby drew attention to the concreteness of a poem, to the very limited difference that a poet makes to the physical environment, to what is going on right there on the page. The important thing to notice is that no writer, no matter how masterful, ever succeeds in expressing an idea directly. What we do, first and foremost, is to arrange words on pages. The amount of pages and words conditions the degree of concreteness that is available to us.

This point bears repeating. The more words you have at your disposal, the more concrete your writing becomes. That is why an "abstract" is so famously short. Some people believe that the abstractness of an academic dissertation results from its length. In fact, however, every page of text, every word (more or less), makes the ideas there more concrete.

Suppose you've got a paper that presents recent empirical evidence that the age of Alfred Chandler's "visible hand" is passing in Bolivia. This paper will, at first pass, be "about" something we can call "Bolivian management practices". If we were to isolate some key words we would say: Bolivia, management practices, Alfred Chandler, Richard Langlois, the visible hand. We can go on from there. Let's give ourselves one-hundred words.

This paper examines current management trends in Bolivia. The Bolivian economy followed international trends after WWII, experiencing its own version of "the managerial revolution", and adopting a largely Chandlerian framework. Like the rest of the world, however, the country is now entering "the new economy" of growing markets. This paper presents empirical evidence to support the suggestion of Richard Langlois that as transportation costs fall and wages increase, the "visible hand" (characterized by the dominance of large, vertically integrated firms) will vanish, leaving smaller, more specialized firms with the competitive advantage.

Given another four-hundred words, we might expand a bit on the Bolivian managerial revolution(s). We could certainly say more about what kinds of firms we have studied. And we can elaborate a little on the views of Richard Langlois and the idea of the "new economy".

Style guidelines often seem to favour these sorts of studies over what is sometimes called "postmodern" writing. I want to show, however, that all this goes just as easily for more "deconstructive" genres. Editors, grammarians and writing instructors ought to be much less despairing about this kind of writing. They ought not, in any case, to despair automatically.

So suppose you've got a paper arguing that Bolivian firms are beholden to the metaphysics presence as the result of the imperialistic influence American management ideologies on Latin America during the Cold War.

Notice, first of all, that the abstraction "Bolivian management practices" captures in three words what this paper is about; that is, these two imaginary papers are at one level of abstraction about the same thing. What they go on to say about that thing is, of course, very different. Key words here might be: Bolivia, management practices, Jacques Derrida, Campbell Jones, metaphysics of presence.

This paper deconstructs the ideologies that currently dominate Bolivian management studies. It targets, in particular, those discourses which owe their dominance to post-WWII (largely American) policy interventions, which are becoming increasingly irrelevant after the collapse of the Soviet empire. As an alternative to uncritically continuing the policies of a bygone era (construed as "texts"), this paper draws on the work of Jacques Derrida to identify the tensions within them . This allows us to note the play of difference that is ultimately the origin of the differences within managerial practice. On the advice of Campbell Jones, it presents this différance both spatially and temporally, indicating both the "elsewhere" and the "later" of management, without succumbing to the "nowhere ever" of utopian approaches.

These two abstracts will allow conference organizers to bring the two papers together or keep them apart as they see fit. There is no simple rule here. The conference organizers may decide to have an interdisciplinary session on current management trends in Bolivia where both papers will serve a useful purpose. Or they may find it more appropriate to run separate sessions on deconstruction and Chandlerian management, each presenting studies of a variety of countries. The point is simply that the abstracts contain enough information to make such decisions intelligently and this is why they should be written.

The abstracts we have written are not inadequate to the ideas that they are about, they are just very abstract. Much more can be said on these topics, and the papers will go on to do so. But once they have been written, they, too, will be "inadequate" to the material that was introduced in the process. There is no ultimate act of concretization. In an important sense, all research writing is the composition of abstracts of the research experience; it is a composite of such abstracts. These are summaries or synopses or outlines of an infinitely rich manifold of impressions that you are helping the reader to make sense of.

Research results are the infinite spaces you can put in a nutshell.

----------
*Here's how David Lehman tells the story: "A century ago in Paris, the painter Degas had lamented that his poems weren't any good though his ideas were wonderful, and the poet Mallarme responded, "But my dear Degas, poems are made of words, not ideas."