4

I was reading a book by Jurafsky et. al. It states following:

  • English adpositions occur before nouns, hence are called prepositions. They can indicate spatial or temporal relations, whether literal (on it, before then, by the house) or metaphorical (on time, with gusto, beside herself), and relations like marking the agent in Hamlet was written by Shakespeare.
  • Subordinating conjunctions are used when one of the elements has some embedded status. For example, the subordinating conjunction that in “I thought that you might like some milk” links the main clause I thought with the subordinate clause you might like some milk. This clause is called subordinate because this entire clause is the “content” of the main verb thought.

Then it gives following table:

enter image description here

Why prepositions and subordinating conjunctions are grouped into the same tag IN in Penn treebank tagset? Is there any linguistic consideration?

1
  • 1
    Unfortunately I don't have a reference for this, but I have heard that some design decisions in the Penn tagset were guided by the ability of the than-available taggers to produce correct tags. It makes the job of a tagger simpler when it doesn't need to disambiguate preposition and subordinating conjunction for a bunch of function words like since, but uses a dictionary and assigns them a strange, but almost always correct tag. Commented Jan 25, 2022 at 17:07

1 Answer 1

1

Look up here:

Complements of P PPs are headed by prepositions and take complements of various categories. The most common are NP and CP, but ADJP, ADVP, IP, PP, etc. are possible as well. Since subordinating conjunctions are treated as P in our corpora, subordinate clauses introduced by a subordinating conjunction are labelled PP. See below for examples. https://www.ling.upenn.edu/hist-corpora/annotation/syn-phrase.htm#pp_comp

2
  • 1
    That just states that subordinating conjunctions are treated as P in the corpora – it doesn’t (as far as I can tell) give any reason for this treatment. Commented Aug 28, 2021 at 12:23
  • If the parser doesn't need the distinction made, why make it?
    – jlawler
    Commented May 26, 2022 at 17:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.