New transitive adjectives

« previous post | next post »

Rodney Huddleston points out to me a remarkable development in English that seems to both of us fairly new (though of course we may be in the grip of the Recency Illusion). English adjectives generally don't take noun phrase (NP) complements. (A complement is a phrase that accompanies a word to make up a phrase having that word as head — for example, something appropriate to a particular adjective that you can add after it to make up an adjective phrase.) The number of exceptions is extraordinarily small: one example is worth (notice how we say worth my time, not *worth of my time). Such exceptional adjectives have long been noted; Fowler comments on worth in his Modern English Usage (1926), and points out that it could be called a transitive adjective. But such adjectives are extremely rare in the dictionary. And yet some new ones appear to have been creeping into the language.

The complements that adjectives do take are mostly preposition-phrases: we say proud of his ancestry, not *proud his ancestry; happy about your success, not *happy your success; familiar with this aircraft, not *familiar this aircraft; and so on. One would expect any new adjectives to follow the usual pattern. But it seems one would be wrong. The adjectives overweight and underweight, in financial contexts of use, have started taking NP complements. In the financial pages you will now find sentences like:

UBS says it is cautious on banks globally, and is underweight European and Japanese banks and neutral on U.S. banks. The broker is overweight Canadian and emerging market banks.

(From this UK Reuters business page.)

Try a Google search on, say {"is underweight the" market}. (Putting market in the set of keywords is a way of fending off spurious hits on weight-watching and dieting sites and largely limiting hits to financial and business material; what we're after is the word sequence "is underweight the", which pretty much forces underweight to be followed by an NP.) What you'll see is pages containing phrases like these:

  • He also is underweight the very long end of the Treasury yield curve
  • the Fund is underweight the Euro
  • and is underweight the South African rand
  • the Portfolio is underweight the US
  • he is underweight the sector in general
  • American Century Investments is underweight the sector
  • Manulife Asset is underweight the U.S. bond market
And so on. You can find examples by the tens of thousands.

What underweight means is basically under-invested. If a bank's principles of investing call for about 25% of its portfolio to be in growth stocks but right now it has only 20%, then growth stocks are underweight in the bank's portfolio. The new usage is to say that the bank is underweight growth stocks.

So there has been a new addition to the extremely select ranks of the NP-complement-taking adjectives. What general conclusion can we draw? The first one that comes to my mind is that it just is not true that languages strive to be regularly patterned and exception-free. They slowly evolve in ways that sometimes level irregularities and sometimes add them; they sometimes increase systematicity of patterning and sometimes decrease it.

Why is that? I have discovered a truly marvelous explanation, which unfortunately the margin of this post is too narrow to contain. (I'm kidding. Nobody knows the explanation. Explaining is one of the missions of the entire field of historical linguistics.)

A second general conclusion is that universal grammar had better not contain a principle from which it rigorously follows that adjectives do not take NP complements. Not only are there a few exceptions, the list actually gets lengthened from time to time. There must be a huge pressure against adjectives taking NP complements in English, but it's not an absolute ban, just a current in the waters of change and probability trending against such a thing. And the current is not strong enough to prevent the language moving a little bit against it sometimes.

Update: 11 minutes after I posted this, Ian Preston emailed mail2languagelog@gmail.com to point out to me that long and short have also been used as transitive adjectives for some time in the financial pages. It would be a very reasonable guess that overweight and underweight (assuming they are newer, which has yet to be confirmed) are following those by analogy. Which tells us something about how linguistic changes of this kind work. Other people (quite a few) are pointing out usages like I'm done supper (Rachel Klippenstein); I was short half a stick of butter (Michael D. Sullivan); I was going to be more but I was shy tens; and (among carpenters) The tenon is proud a sixteenth (meaning "The tenon sticks out by 1/16 of an inch"; Dan Lufkin).

Incidentally, I am well aware that you might think of long the dollar as being a shortened version of "long in the dollar", and done supper a shortened version of done with supper, and proud a sixteenth a shortened version of proud by a sixteenth, and so on. Certainly one could look at it thus. But calling it leaving out a preposition doesn't put us any closer to understanding why this particular preposition can be left out in this context; there is a syntactic change under way here, and it's leading to apparently new cases of transitive adjectives, and I'm trying to identify where, and with which items.

One other thing that people are writing to me about is that they think worth might be a preposition rather than an adjective. This has been argued before. I think it's wrong. For the syntacticians among you, let me just say (i) try pied-piping it, and (ii) try using it as the head of a fronted loose adjunct and see whether it needs something to be predicated of (it does). Check chapters 6 and 7 of The Cambridge Grammar of the English Language for the relevant tests.

One more update: Sasha Albertini has heard people say things like I'm good at least ten bucks, meaning "I have at least ten dollars more than I need". So that would mean that good (pretty much the archetypal adjective, the one that is found even in languages that have hardly any members of the adjective class) has picked up a transitive use as well.

[Comments are closed because I'm overweight comments at the moment. But you can mail me at mail2languagelog@gmail.com if you like, and I will pay attention.]



Comments are closed.