Friday, August 07, 2009

A newly discovered preposition

I believe that I may be the first person to have realized that compared is a preposition. It is not listed as such in any of the dictionaries that I consulted, and you may very well be wondering how compared could possibly be a preposition. Let me try to explain.

The obvious alternative is that it is a verb, a past participle to be precise. One property of participle verbs is that, when they function as adjuncts, they take the subject of the main clause as their subject. When people overlook the requirements, you end up with a silliness like the following (from here):
1 Now 83, and long gone from power, Britons remain fiercely divided over the reign of former Prime Minister Margaret Thatcher.

Notice that we expect this to mean that Britons are 83 and long gone from power, because Britons is the subject of the main clause. When the main clause has a dummy subject, like there, again we end up with problems (from here).
2 Fearing a massive lay-off, there was a general sense of relief when the boss announced there would be no new budget cuts.
But when we replace the participle with a preposition, everything is fine.
3 After a massive lay-off, there was a general sense of anger.
Our second piece of the argument is that, despite traditional thinking to the contrary, prepositions do not actually need to come before nouns. They can come before other prepositions, as they do here:
4 She came in from out of nowhere.
Now, we can understand why compared is a preposition. Consider the following sentence:
5 Compared to ICS alone, there was a significantly greater improvement in FEV1 with the addition of LABA.
To be honest, I don't know what it means, but it seems to me that there is no problem here analogous to what we faced in 2 even though we have a dummy subject. That means compared is not a verb. Perhaps, then, it could be a preposition along the lines of after in 3 and followed by another preposition to as in 4.

There are other prepositions like this:
  1. according
  2. allowing
  3. barring
  4. compared
  5. concerning
  6. considering
  7. counting
  8. depending
  9. excepting
  10. excluding
  11. failing
  12. following
  13. given
  14. granted
  15. including
  16. owing
  17. pending
  18. pertaining
  19. provided
  20. providing
  21. regarding
  22. respecting
  23. saving
  24. seeing
  25. supposing
  26. touching
  27. wanting
Some of these are recognized by major dictionaries. The Longman Dictionary of Contemporary English lists given as a preposition. And the online OED has barring, concerning, and considering among others. None, however, has the full list above, except the Simple English Wiktionary, (which by the way now has headword entries for the most common 2,000 words of English, the Basic English 850 list, and the Academic word list. All told, it now has over 10,000 entries.)

15 comments:

  1. Traditional grammar would say that "compared to" is the preposition — likewise "according to",
    "owing to", and "pertaining to". Is there any reason to think of "compared", "according", "owing", and "pertaining" themselves as prepositions? (Is this like intransitive prepositions, but instead, prepositions that construe their complements with "to"? If so, it's no wonder that major dictionaries don't have them, since they don't believe in these kinds of prepositions.)

    ReplyDelete
  2. Yes, it is like intransitive prepositions. But there is no reason to believe that "compared to" is a unit. The same works with "compared with". Also, the verb is "compare" not "compare to", so why would the preposition be "compared to"?

    ReplyDelete
  3. Anonymous6:34 AM

    I'm not convinced. It seems to me a rather flimsy basis for distinguishing parts of speech, and it doesn't help that there are number of important differences between (2) and (5).

    That said, I agree with your point in the "Slippery Parts of Speech" post that boundaries between parts of speech are often arbitrary. To make that point another way: theoretically, you could model the language with a giant Venn diagram, with one circle representing, for example, the set of words that are grammatical in the context of "Theoretically, you could _____ the language with a giant Venn diagram". But this is impractical, and parts of speech facilitate discussion of the language in terms of patterns and tendencies, e.g. that a word with these properties will usually have these properties also.

    ReplyDelete
  4. What basis, outerhoard, do you find flimsy? On what basis, do you think the OED has 'barring', 'concerning', and the others as prepositions?

    'Slippery parts of speech' was posted by Q, not Brett. I take Q's point to be that it's all about the analysis rather than about putting things in the right slots (though I think he may have bee taking a gentle poke at me).

    ReplyDelete
  5. Anonymous3:12 PM

    Will the madness never cease? Geoff Pullum seems to have started it with his claim that bush is a preposition here in his interview on ABC's Lingua Franca. The examples he uses to attempt to establish the category of the intransitive preposition are clearly prepositional phrases with the object understood. Bush in his example is clearly the object of an understood preposition. Any attempts to classify bush as an adverb are similarly misguided. The prepositional phrase (to the) bush is acting adverbially but the object, bush is not itself an adverb. In the case of compared any attempts to cram it into the prepositional pigeonhole result from a failure to understand that English grammar is gaining the category of topic, which, like in Japanese, may or may not be the same as the subject. The so-called dangling participles usually can also be resolved by seeing them as referring to the topic of the sentence.

    Faldone (whose password doesn't seem to be working)

    ReplyDelete
  6. Faldone, could you explain why this category of topic applies only to some participles?

    ReplyDelete
  7. Sometimes the participle isn't dangling and sometimes it's simply a case of a very poorly constructed sentence. I'm just saying that the so-called "dangling participle" isn't always bad grammar, just inadequately described grammar. I'm more concerned with this torturing of the definition of preposition.

    ReplyDelete
  8. Faldone, could you give some examples? It seems to me that it is a very small list (limited mainly, if not entirely, to the participles that I listed) that can be grammatically "dangled". I don't see how that squares with an idea that English is growing a topic category. Perhaps you'd like to write a guest post and elaborate?

    ReplyDelete
  9. I'm sorry I got side-tracked into the dangling participle issue. I'm really interested in these prepositions. Can anyone give me a definition of preposition that includes the Pullum-Huddleston intransitive prepositions, particles of phrasal verbs, and traditional prepositions?

    ReplyDelete
  10. I don't think any simple definition is going to cut it, but their definition from A Student's Introduction to English Grammar defines prepositions as "a category of words whose most prototypical members denote relations in space or time (in, on, under, before, etc.) and take NPs as complement (in the car, on the chair)." Other characteristics which are typical, though not universal include the ability to function as a non-predicate adjunct and to be modified by right and straight, and the inability to function as predicative complement to become.

    ReplyDelete
  11. Hi Brett,

    No, I wasn't taking a poke at you, or at anyone else, really. I was perhaps lamenting over the kind of linguistic dogmatism where people tend to impose a fixed view (often based on their pet theory) on linguistic phenomena.

    Hi Faldone,

    While I do see your point, let me point out that CGEL proposes and provokes, rather than decides and dictates. Sure, it is a grammar book and it says something like 'we treat X as a preposition'. But what it means is that 'under the sort of analysis we adopt here, we are forced to say X is a preposition. If you are not happy with it, fine (but you still have to explain the relevant facts ...)'. Take a quick look around page 19 of CGEL.

    ReplyDelete
  12. Anonymous12:56 PM

    Hi Q,

    Faldone here. Just to take one category, the intransitive preposition; I don't think I've seen one example of this proposed category where the object wasn't assumed. Even in the case of "afterwards", which probably never takes an explicit object, the object is implied. If I were to send you an email that simply said, "I'll meet you at Joe's Bar & Grill afterwards," your response would be something along the lines of "afterwards of what?"

    ReplyDelete
  13. Hi Faldone,

    Your comments could be construed as an argument for analysing afterwards as a preposition (or an 'intransitive' preposition) - albeit on a semantic basis. But then, we cannot arrive at a reasonable and solid conclusion by accepting something 'implied' as grammatical evidence. CGEL (p.616) admits that the case for analysing afterwards as preposition is a little weak.

    The bottom line is, afterwards is not easy to deal with, and the difficulties will not go away if I mumble something like 'adverb', unless I clearly state what that is.

    BTW, one way to feel better about these is an etymological approach. With afterwards, for instance, its etymological sense/construction is 'after-something-ward', where the 'something' provides a slot for whatever is implied in I'll meet you at Joe's afterwards.

    Still not feeling better? Well, that's not my fault; it's the language.

    ReplyDelete
  14. Anonymous8:16 AM

    Just wish my browsers would decide whether to let me be Faldone or not.

    If we could apply Gödel's Incompleteness Theorem we could say that language is not capable of describing language.

    ReplyDelete
  15. In a way, that's quite true: language describing language embodies a fundamental problem. Linguists usually put this aside by calling the language employed to describe a language 'metalanguage' (accordingly, in order to describe a metalanguage, one needs a metametalanguage to the language; hey, I am not kidding: take a look at page 54, footnote 1, of Syntactic structures by Chomsky to see the word in print).

    ReplyDelete