I am a little puzzled. Is there any such thing as an A-position where no structural Case is available? The arguments are either a nominative or an accusative (structural Cases) aren't they?

[...] Since this problem arises from the fact that [+F]NPs can occur in A-positions where no structural Case is available, [...] (Kobayashi, K. (1987). A note on bare-NP adverbs. English Linguistics, 4, 336-341.)

1 Answer 1


There are A-positions where no structural case is assigned. Let me give you some examples from English:

(1) The data were classified
(2) They arrived
(3) They seem to have arrived
(4) The data seem to be classified

(1) is a passive sentence. The complement position (V [DP]) is an A-position where no structural accusative case has been assigned. This position is left without its case (or say the passivized verb does not assign accusative case), this is why the object the data has to move to another position where it ends up assigned a nominative. A similar observation carries over to (2) as an unaccusative verb.

The sentences in (3)/(4) are examples of raising to subject. The specifier position (A-position) of an embedded infinitival clause does not assign a nominative case. Again their subjects have to move further in the structure ending up assigned a nominative by a tensed matrix clause.

Notice that (3) combines two A-positions left without case. The first is the embedded A-position (V-complement) of the unaccusative arrive and the second is the embedded infinitival position (Spec,IP). The same thing with (4), there are two positions left. The embedded passive V-complement and the embedded infinitival Spec,IP.

These are the two structural case positions left without case in a language like English. This observation, however, varies from one language to another.

  • Thanks, Tsutsu. Okay, so the author was talking in perspective of D-structures. I was thinking in terms of only S-structures. He went on to say, " (... A-positions where no structural Case is available,) one might try to solve this problem by excluding the activation of the [+F]feature from A-positions, i.e., by restricting it to A'-positions or adjunct positions." Would A'-positions be A-positions where no structural Case was assigned?
    – Sssamy
    May 21, 2020 at 12:47
  • Case is assigned at S-structure in GB framework. A'-positions (or adjunct positions) are not assigned case. But elements which were assigned case before may move there, e.g., wh-movement. A' positions would never be A-positions, but A'positions can host elements which were assigned case in A-positions.
    – Tsutsu
    May 21, 2020 at 13:51
  • I stand to be corrected: Since V [DP] is in the representation of D-structure as you can tell it from seeing the fact [DP] is moved to the subject position (at S-structure), the A-position you said initially is in D-structure. Correct?
    – Sssamy
    May 21, 2020 at 22:51
  • Sorry, Tsutsu, one more thing. What are you referring to when you say 'elements which "were" assigned case "before"'? I'm a non-native speaker. I couldn't get it. Thanks.
    – Sssamy
    May 22, 2020 at 0:15
  • I might have had to say 'What were you referring to when you said ...?'
    – Sssamy
    May 22, 2020 at 0:25

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.