The Negation in 'Do Not Believe' Statements

[iamtaka]

Atheists often claim, "I do not believe God exists." When asked to clarify what they mean, they point out that they lack belief in God's existence. When they read the sentence provided, they see the negation as being applied to the verb believe and not the content of belief (i.e., God exists). That is, "do not believe" means simply "lacking belief."

Alternatively, if the negation applied to the content of belief, then the statement could be rephrased as "I believe no God exists." or "I believe God does not exist." This means the atheist would not "lack belief in God's existence" but would rather hold a belief about the non-existence of God.

Question for debate:
Q. Within a linguistic context, does the negation apply to the verb believe or the content of belief?

[Nickman]

I know many atheists who hold to the idea that "I do not believe that God exists" and "I believe God does not exist" are different. They are in fact the same statement and claim.

Hang on a tick, Nick.

The first phrase could be said by an agnostic, the second one could not; so there is a difference somewhere.

Both sentences are the same. If a person does not believe that God exists, they are saying that they don't believe that God exists. The agnostic says we cannot know. An agnostic doesn't say either way. They make no claim. If you say you do not believe or you do believe, you have made a soft claim. One that doesn't require support. The only claim that an agnostic makes is that we cannot know either way.

[Jax Agnesson]

What are you trying to get at if a person admits that belief is involved?

. . . .. I am . . . demonstrating that negation is applied semantically/pragmatically to the object.

Consider the following exchange:
Q: "Do you always believe what you read in the papers?"
A: "No."
This negation attaches itself more to the belief than to the unspecified editorial content.

Now compare that with
Q: "Do you believe everything you read in the papers?"
A: "No."
This negation surely is attached to the 'everything' under question, ie to the content, or some part of it. In each case, the emphasis is directed by the questioner, through the placement of the qualifiers 'always' and 'everything'.
Perhaps this hints at a way to approach the a/theism angle.
'Do you believe in (some particular notion of) God?' surely invites a reference to the object, ie some specific God (eg Ganesh or Allah.)
'Do you have a firm personal belief in (an unspecified) God?' invites a reference to faith itself.

The lesson surely is that the questioner needs to be aware of the extent to which she, consciously or otherwise, directs the response; and a statement of non-belief or dis-belief cannot be understood simplistically, relying only on a preconceived notion of the rules of grammar.
IOW, if you ask me whether I believe in God and I answer 'No', you are not yet in a position to tell me what my answer means.

[iamtaka]

So if someone asks me 'Do you believe that x?' and I answer 'No.' how does that fit into your liguistic schema?

No change. It's still a semantic/pragmatic issue. The simple response of "No." incorporates the context. That is, pragmatically, we understand "No." to be equivalent to "I do not believe that x."

Normally, I would expect the negator to attach to the verb. 'No, I don't believe that x'.

Is this connected to the discussion which followed? I was unclear whether it was a standalone statement or connected to something.

But if someone asks 'Is x the case?' I can either answer 'No' (as if stating a fact) or I can answer 'I don't believe so'. (as if stating an opinion or estimation.)
The first option attaches the negator to the object 'x', the second to the belief.[/ The really significant difference is that the first option 'No.' does not make explicit that knowledge is a degree of belief.
Again we see the flexibility of modern English. We can apply the negator either to the belief or to the object of belief, giving exactly the same information factually, but allowing a range of nuance.

First, 'believe' is part of a category of verbs called negative raising predicates. 'Be' is not. This means there are semantic and pragmatic differences between the two examples questions. In other words, it's an apple and oranges comparison.

Second, the flexibility which you're observing has to do with a pragmatic shift not negation.

'I do not believe that x' and 'I believe that not x' are subtly different in usage, but semantically they are equivalent, to the extent that either can be correctly logically deduced from an exchange such as:
Q: 'Do you believe that God exists?'
A: 'No.'

I agree with you here.

[iamtaka]

I understand this because I agree with you, but what is the further, hidden goal of your post? If you establish that atheists have belief, that means nothing, unless you have a further goal.

There is no hidden goal. My goal is not to establish that atheists have belief, but to establish that a subset of atheists misappropriate English in defense of a belief about a definition.

[Nickman]

I understand this because I agree with you, but what is the further, hidden goal of your post? If you establish that atheists have belief, that means nothing, unless you have a further goal.

There is no hidden goal. My goal is not to establish that atheists have belief, but to establish that a subset of atheists misappropriate English in defense of a belief about a definition.

So what does that accomplish? We all have beliefs. Some have more evidence to support them.

[keithprosser3]

Both sentences are the same. If a person does not believe that God exists, they are saying that they don't believe that God exists. The agnostic says we cannot know. An agnostic doesn't say either way. They make no claim. If you say you do not believe or you do believe, you have made a soft claim. One that doesn't require support. The only claim that an agnostic makes is that we cannot know either way.

1 - "I do not believe that God exists"
2 - "I believe God does not exist"

Sorry to nitpick, Nick, but let me repeat; an agnostic can say 1) but not 2) so the phrases cannot be precisely the same in meaning.

But what significant point remains of this thread has me baffled.

[Jax Agnesson]

I know many atheists who hold to the idea that "I do not believe that God exists" and "I believe God does not exist" are different. They are in fact the same statement and claim.

Hang on a tick, Nick.

The first phrase could be said by an agnostic, the second one could not; so there is a difference somewhere.

Both sentences are the same. If a person does not believe that God exists, they are saying that they don't believe that God exists. The agnostic says we cannot know. An agnostic doesn't say either way. They make no claim. If you say you do not believe or you do believe, you have made a soft claim. One that doesn't require support. The only claim that an agnostic makes is that we cannot know either way.

Suppose you ask me whether I have a fondness for haggis, and I answer 'No'. This could be for one of two reasons:
a: I have tried haggis and didn't like it, or
b: I've never tasted it so haven't had the chance to develop any such fondness.
So there are two very different sets: the people who dislike haggis, and those who have never tried it.
( I am told there may be a third set, of people who actually do like haggis. But I rule out that possibility! )

'I lack a liking for haggis' and 'I have a dislike for haggis' are not the same.
'I lack a belief in God.' and 'I have a disbelief in God' are also different.

[iamtaka]

Consider the following exchange:
Q: "Do you always believe what you read in the papers?"
A: "No."
This negation attaches itself more to the belief than to the unspecified editorial content.

Now compare that with
Q: "Do you believe everything you read in the papers?"
A: "No."
This negation surely is attached to the 'everything' under question, ie to the content, or some part of it. In each case, the emphasis is directed by the questioner, through the placement of the qualifiers 'always' and 'everything'.

To put it crudely, adverbs are going to change the grammaticality of a verb.

I don't have time for a fuller response at the moment. If you want some more in-depth reading on the subject, you can see Gajewski on NRPs and Edlestein on Adverb Climbing.

IOW, if you ask me whether I believe in God and I answer 'No', you are not yet in a position to tell me what my answer means.

True, but I am in a position to tell you what it means to me.

[iamtaka]

So what does that accomplish? We all have beliefs. Some have more evidence to support them.

It's a baby step.

[iamtaka]

1 - "I do not believe that God exists"
2 - "I believe God does not exist"

Sorry to nitpick, Nick, but let me repeat; an agnostic can say 1) but not 2) so the phrases cannot be precisely the same in meaning.

See the Gajewski paper to which I linked above.