Adjectives and past participles

Returning to Leo Schmitt's most recent Grammatically Speaking column, I'd like to take up the issue of when a past participle becomes an adjective. This point comes up in the "Brain Teaser" section:

The following two sentences are grammatical. They both demonstrate the same exception to a general grammatical rule of English. What is that rule and what is the exception?

1. The Natchez Board of Aldermen last week voted unanimously to demolish the collapsed building at the corner of Martin Luther King Jr. and Franklin streets. (From “City to demolish collapsed building in downtown,” The Natchez Democrat.)

2. Authorities believe an escaped convict on work release ... is responsible for the blaze. (From “Investigators believe escaped convict set Roan Mountain fire,” www.tricities.com.)

The rule to which these sentences appear as exceptions is the rule that when a past participle is used as an adjective it should have passive meaning; in other words, the noun should have the role of “done-to” rather than “doer” with relation to the event denoted by the participle. This is the case in broken window, for example. In the examples above, by contrast, the noun has the role of “doer”: the building has collapsed, and the convict has escaped. Other similar examples include fallen trees and elapsed time.

Jeffrey Rasch, Denton, Texas

Excellent job, Jeffrey. That is exactly the answer I was looking for. You are right that in general, we expect adjectives derived from past participles to have a passive sense. This is connected to their uses in passive constructions. Thus the example you give, broken window, can easily be the window that was broken and this is the general grammatical rule. There are, however, a comparatively small number of adjectives derived from past participles such as the example you gave, where the sense is more of an agent or “doer,” so we say the escaped convict but NOT the convict who was escaped. Many thanks.

In his answer, Rasch talks about a past participle "used as an adjective," where in his explanation, Schmitt refers to "adjectives derived from past participles." I would argue that the first is the rampant category-function confusion, while the second is accurate at times, but goes too far in this case.

First, we need to distinguish between lexical categories (or "parts of speech", e.g., NOUN, ADJECTIVE, PREPOSITION, etc.) and functions (e.g., SUBJECT, COMPLEMENT, MODIFIER, etc.). Adjectives perform a variety of functions. These include attributive modifier in NPs (e.g., the big Lebowski), complement in prepositional phrases (e.g., left him for dead), adjunct in clauses structure (e.g., Happy with the outcome, they headed for home), and others. But ADJECTIVE itself is not a function. It makes no sense, then, to say that a past participle is "used as an adjective." Rather, what Rasch should have said is "when a past participle is used as an attributive modifier in an NP."

We could, perhaps, take "used as an adjective" as shorthand, but it is shorthand that leads to confusion. This can be seen in Schmitt's summary. He's gone all the way to putting the past participles in the ADJECTIVE category. It is indeed the case that some past participles have become adjectives, but that's not the case for most, and certainly not with escaped and collapsed.

Examples of past participles that have truly become adjectives are interested, tired, and bored. Note that all three can be modified by very and that they can perform all the normal functions of adjectives. It is fine to say she became interested in the subject, but not to say *the building became collapsed.

Again, the correct analysis is that past (and present) participles are verbs while accepting that some have become adjectives. Either way, it makes no sense to say they are "used as" or "function as adjectives", but they do function as attributive modifiers in NPs. If we want to distinguish between those that also belong to the adjective category and those that don't, the very test is useful, as is the become test.