Dave's ESL Cafe's Student Discussion Forums

[senry]

hi. Here is a question I would like to ask you:

1 He didn't think that Mary was pretty.
2 He thought that Mary wasn't pretty.

I know these two sentences are semantically equivalent. But
why are verbs in similar semantic field of beliefs, such as assume and surmise treated differently?


3 I don't assume that he came.
4 I assume that he didn't came.

according to a reference book, these two sentences are not equivalent, but what difference does it make
semantically?

thank you in advance,
senry
_________________
hi. I am keen on learning languages and I would be greatful if any of you might answer my questions. cheers.

[Anuradha Chepur]

This belongs to the area of hardcore theoretical linguistics - syntax and semantics. It has to do with negation scope and truth value semantics. I do not know if you are into studies of that kind. So I would prefer to give a simple non-technical explanation:

Transfer of negation from the main to the subordinate clause is allowed only in the 'believe' type verbs, as they are called and 'think' is one such as is 'consider'. But 'assume' isn't. So you get equivalent readings for your 1 and 2, but not for 3 and 4.

[2006]

4 should be, "I assume that he didn't come."

[Anuradha Chepur]

Corrigendum:

In my last post, which was typed in a hurry, what I wanted to say was truth-conditional semantics and not truth-value semantics.

[senry]

Hi Anuradha.
Thanks for your answer.
Just out of curiosity, could you explain a bit using the framework of the truth-conditional semantics? If it is like mathematical categorization, I would not be able to follow, though.
_________________
hi. I am keen on learning languages and I would be greatful if any of you might answer my questions. cheers.

[Anuradha Chepur]

Hi Senry,

I would tentatively analyse it this way: For instance, for the factive know-type verbs (learn, regret) the main proposition presupposes the embedded proposition. If X is true, then Y is also true.

1. John knows that Mary is not beautiful. (True)
2. Mary is not beautiful. (True)
3. John doesn�t know that Mary is beautiful. (True)
4. Mary is beautiful. (True)

#1 presupposes #2. The proposition that John knows that Mary is not beautiful presupposes that Mary is not beautiful. . If #1 is true, then #2 is also true. Similarly #3 presupposes #4. If #3 is true, then #4 is true, so �Mary is beautiful� is true. Therefore #1 and #3 are not semantically equivalent.


This is not so for the non-factive believe-type verbs. (think, consider, expect, seem)

5. John believes that Mary is not beautiful. (True)
6. Mary is not beautiful. (True/not True)
7. John doesn�t believe that Mary is beautiful. (True)
8. Mary is beautiful. (True/not True)


If #5 is true, it doesn�t follow that #6 is true. #6 may or may not be true. Similarly if #7 is true, it doesn�t follow that #8 is true. The actual state of affairs may be different.

The same truth conditions/values apply for volitional claim-type verbs. (wish, assume etc), so we tend to think that assume is the same class as think, but it isn�t.

9. John assumed that Mary is not beautiful. (True)
10. Mary is not beautiful. (True/not True)
11. John didn�t assume that Mary is beautiful. (True)
12. Mary is beautiful. (True/not True)


However, only believe-type verbs allow negation scope of the main clause over the embedded clause and not claim-type verbs. This is because both have different semantic and syntactic properties.
i. Believe-type verbs, it is said, do not allow independent truth value to the embedded clauses. The whole sentence has to be treated as a single truth value domain.
ii. Only believe-type verbs allow neg-raising from the embedded to the main clause (or lowering from the main to the embedded clause as it is debated).
iii. Further, believe-type verbs allow for sentential-negation, i.e., the negation is adjoined to the sentence, whereas other verbs allow only constituent-negation, i.e., the negation is adjoined only to the verb.

There are a lot of (inconclusive) studies on this and the issues are still open to debate.

[senry]

Thanks again Anuradha. I really appreciate the detailed exlanation.
I have been thinking about this since then.
I do not follow all of what you wrote here but at least I can understand
a part of the logic behind it. Maybe I should read books on it if I would like to know further.
I would like to ask you another question in the near future if I may.
Thanks again.

henry
_________________
hi. I am keen on learning languages and I would be greatful if any of you might answer my questions. cheers.

[Anuradha Chepur]

You are welcome Senry. You may please ask further questions. If I may know what exactly you didn't understand, I can explain further or suggest the relevant reading.