The http://lindat.mff.cuni.cz/services/udpipe/ statistical parser gives me two different analyses (one with ‘finished’ as a verb and the other as an adjective). ERG 2018 and trunk give me finished as VERB. I know that this form can be ADJ or VERB. But in this particular case, WHY the VERB analysis was taken? For I am tired I have three analyses (verb, noun, adj) … For I am exhausted I got two analyses (adj and verb).
Does the verb ‘finish’ have any particular property compared to exhaust?
It is also interesting that google translate gives me two different translations:
I’m finished => Eu terminei
I am finished => Eu estou acabado
The first is the VERB reading, I have finish and action, finish as a intransitive verb in the simple past. The second is something like: I am in a particular condition of begin finished (in Portuguese it would be like tired or exhausted). But the BE + -ED is a not a English construction for simple past… so the first translation is wrong!
I think the statistical parser is generating the two possible analyses for the two example sentences. The sequence Subject + BE + Past Participle may represent two different constructions in English:
(1) the passive construction
(2) the copulative predicative construction
In (1), the participial form is a verb, while in (2) it should be analysed as an adjective. (1) is licensed by transitive verbs, i.e., verbs governing a direct object. (2) is licensed by telic verbs, cf.:
(3) *the dog was barked (bark is intransitive, no passive reading available)
(4) *the woman is danced (dance is non-telic)
The verb finish has many readings, among which the transitive one. This explains the availability of the passive reading for the example sentences. I am not aware of a noun reading for tired, but it may exist.
In the online Cambridge dictionary, exhaust is only assigned a transitive valency frame. Since this verb is also telic, both (1) and (2) are possible readings for I am exhausted.
I think both translations are possible, depending on the underlying reading of the corresponding verb, which, in turn, depends on the larger discourse context, but I suggest a native speaker should be consulted. (1) may express the situation of having completed the action of having finished something, e.g., a task, cf. A1 in the Cambridge dictionary entry, (2) expresses the result of having being finished. A third reading would be the passive one, cf I am finished by someone, i.e., the process of being finished. This reading may seem weird. A possible context would be the action of an artist which has painted the picture of someone. In this case, the subject I refers metonymically to the painting, i.e., “my painting is completed”.
I think the ERG is missing a lexical entry for “finished”. For me, this adjective is definitely polysemous, so “I’m finished” could be be paraphrased as “I’ve finished” or “I’m exhausted” or “I’m done for”, depending on the context. I have a vague sense that “I’m finished” is more informal than “I’ve finished”. It can also take a with-PP complement, as in “I’m finished with my work”. I think it can also be used attributively, although this might be less common. In the context of running a race, “the finished participants” could refer to the participants who have finished the race.
For me, this is more lexically restricted than all telic verbs, e.g. I wouldn’t use “completed” in the same way. But a quick Google search suggests that it is sometimes used as an adjective (e.g. “the completed participants” ~ “the finished participants”). So there may be some dialect or idiolect variation here.
I guess the next obvious question is how to deal with such situations where ERG is missing a lexical entry.
@Dan have already helped me in the past with particular situations, that is, he gave me fragments of TDL with specific instructions about where to put them. I also remember some advices about WHEN adding such new entries make sense. As far as I understood, his suggestion is to be lazy, letting the guesser do the job whenever possible and adding entries only then the guesser is making wrong decisions. The point was that once in the lexicon, the guesser will be ignored and we will not be able to deal with different POS/lemma for different contexts. So the lexicon will have to anticipate all possible interpretations.
But I didn’t understand how to find the right lexical type and create the TDL entries by myself. For the verb finish I found
But they are almost encrypted to me! Fortunately, it looks like the most important decision is what super type the entry must have, since ORTH and SYNSEM seems obvious. I can’t say the same about PHON.ONSET and --COMPKEY. It looks like one needs to understand the lexicon organization in ERG to make any change on it.
I believe PHON.ONSET is used to determine whether to use “a” or “an” as the indefinite article (and maybe other things?). con means consonant, so it would be “a finished race” versus, e.g., “an ongoing race”.
Working from most recent backward in this thread:
Entries in the ERG’s “lexicon.tdl” file must each provide at minimum the name of the entry, its lexical type, its orthography, its semantic predicate (except in rare instances), and its PHON phonological onset (vocalic or consonantal), as Mike mentioned, so the grammar knows whether this word might be preceded by “a” or “an” (as in “an honest mistake”). For verbs, adjectives, and nouns that take a complement, the lexical entry may also specify some idiosyncratic property of that complement, via the --COMPKEY or --OCOMPKEY (oblique complement) attribute, typically for verbs that select for a specific particle or prepositional phrase; for example, the verb “look” used in “look up the answer” is of type v_p-np_le and requires its first complement to be “up” via its --COMPKEY value.
Users of the ERG are of course welcome to add their own entries to the lexicon provided, but need to remember that if one adds a new entry, then all words with that same stem must also be defined, since the POS-tag-based handling of unknown words is disabled for any stem that has at least one entry in the manually-defined lexicon.
Now on the main question about the lexical ambiguity for “finished” and similar words: I agree with Alexandre and Guy that the ERG is currently missing an entry for “finished” as an adjective. Apart from often fuzzy intuitions about word senses, there are a few syntactic tests to distinguish adjectives from passivized verbs, and for at least one of these tests (combining with “… enough to…”), “finished” can be an adjective: “This article is finished enough to submit as a short paper”, though it for some reason does not like the degree specifier “very”, another usually reliable test for adjectives: “?This article is not very finished yet”.
It would nice to have some systematic way of discovering a relatively complete inventory of adjectives that have the surface form of a passivized verb; the list of ones included in the ERG lexicon is, I fear, far from complete.
Fortunately, it looks like the most important decision is what super type the entry must have, since ORTH and SYNSEM seems obvious. I can’t say the same about