The Coherence of Naturalism (Atheism)

[Haven]

One of the major criticisms of metaphysical naturalism (and by extension, atheism) is that it is logically inconsistent, with some even going as far as to say that it is self-undermining. This criticism, which comes primarily from Reformed Christian thinkers such as Cornelius Van Til, Alvin Plantinga, John Frame, and our own theopoesis, is due to the fact that (according to Reformed thinkers) naturalism cannot account for several features of human existence, such as (objective) morality, existential meaning, consciousness, free will, and aesthetic taste.

Plantinga goes further, saying that a combinstion of evolution and naturalism would lead us to develop cognitive faculties geared for survival rather than truth, meaning that, on naturalism, we would be unable to know whether or not any of our beliefs -- including naturalism itself -- are true, meaning naturalism defeats itself.

As an alternative to naturalism, Reformed thinkers believe Christian theism can account for true cognitive faculties, teleology, morality, beauty, etc., and that it should therefore be preferred over atheistic naturalism. They use a presuppositional approach to illustrate this, arguing that Christian presuppositions are required for a coherent worldview.

Atheists, in my experience, rarely respond to these criticisms. When they (we) do, they tend to defend a naturalistic account of cognitive reliability while writing off morality, aesthetics, knowledge, etc. as subjective or illusory. Naturalistic thinkers also tend to point toward philosophical problems with theism, such as Michael Tooley's (2008) update of the problem of natural evil. Additionally, one major naturalistic response to this comes from philosopher Feng Ye, who attempts to give a naturalistic account of cognitive reliability. The book "Naturalism Defeated?" (2001) was written in response to Plantinga's argument.

Debate question: Are the Reformed thinkers right? Is naturalism coherent? Can atheists account for morality, purpose, etc. on naturalism? Are Christian presuppositions necessary for a coherent worldview? Does Plantinga's argument succeed? Is theism coherent?

[Furrowed Brow]

I fail to see how my use of the word "external" is relevant to your critique of my post. Empirical observation is by definition external to the proposition it supports, therefore, I don't see how I misspoke.

The problem is that the word external introduces the potential to think Ayers verification invokes a mind (internal) world (external) distinction, or some similar ontological distinction. Here Ayers verificationism is neutral and there is no such commitment. Ayer fully formulates his position across several chapters. He is not committed to the distinction of external your formulation leaves him open to, other than in the weakest sense that the possibility of the sense-experience is not reliant on the proposition. For instance, claiming there is heaven, and then claiming that we will experience heaven when we get there, fails the verification principle because the possibility of the sense-experience in this example is dependent on the truth of the proposition. If this is what you meant by external then Ok, but I did not read it that way in the original context. Maybe the possibility of the empirical observation is independent to the proposition it supports would serve your point better.

Which still leaves the principle self-stultifying: the verification principle is not an analytical proposition, and would itself require verification.

I suspect Ayer think he is doing philosophy and introducing a logical proposition and therefore an analytical proposition.

I maintain in this book that it is not within the province of philosophy to justify our scientific or common-sense beliefs; for their validity is an empirical matter, which cannot be settled by a priori means

So if ayer means verificationism to be a philosophical concept then it is not intended to be an empirical proposition. Ayer gets there by an analysis of what it means for a proposition to have factual content. In the Ayer quote you provide he demands that an empirical hypothesis be supported by at least the possibility of some sense experience by which the empirical hypothesise might be determined to be true or false. This looks like a tautological truth of what it means to be an empirical hypothesis, and thus is an analytical proposition. The evidence you provided does not support the point you are making and serves to support the exact opposite. To counter you simply have to show where Ayer fails to form a tautology.

This principle would need to be applied to itself, and I don't see how it could pass its own test. What empirical observation leads one to believe that factual propositions require the support of some empirical observation?

It runs like this " propositions of matter of fact are empirical hypothesis and can only be such if there is the possibility of some sense experience for determining whether they are true or false. So if there is no possibility of sense-experience by which to determine the truth or falsity of a statement, the statement is not empirical. The logic is hard to deny and not self refuting in any obvious way.

[Haven]

I suspect Ayer think he is doing philosophy and introducing a logical proposition and therefore an analytical proposition.

I don't see how the verification principle meets the definition of an analytical proposition, which is defined by Irving Copi (1953) as:

. . . [a proposition] which is necessarily true because its denial involves a contradiction.

The quote is from the article Analytic Philosophy and Analytical Propositions, appearing in a 1953 volume of the journal "Philosophical Studies," pp. 87.

This definition is also reiterated by the Stanford Encyclopedia of Philosophy, which states:

Analytic sentences, such as Ophthalmologists are doctors, are those whose truth seems to be knowable by knowing the meanings of the constituent words alone . . .

Given these definitions, the verification principle does not meet the definition of an analytic proposition, and is therefore self-stultifying given my earlier comments.

However, you insist that the principle is tautological and therefore analytic, saying:

In the Ayer quote you provide he demands that an empirical hypothesis be supported by at least the possibility of some sense experience by which the empirical hypothesise might be determined to be true or false. This looks like a tautological truth of what it means to be an empirical hypothesis, and thus is an analytical proposition. The evidence you provided does not support the point you are making and serves to support the exact opposite. To counter you simply have to show where Ayer fails to form a tautology.

Sure . . . in the Ayer quote I provided, he said:

I require of an empirical hypothesis not that it should be conclusively verifiable, but that some possible sense experience should be relevant to the determination of its truth or falsehood.

I feel you equivocated on the meaning of the word "empirical" as it concerns Ayer's quote. In the first sense, "empirical hypothesis" seems to suggest any non-analytical, non-tautologous factual proposition, for example, the statement "God exists." In the second sense, Ayer refers to empirical evidence that could be used to justify such a proposition. This is not tautologous.

If a putative proposition fails to satisfy this principle, and is not a tautology, then I hold that it is metaphysical and that it is neither true nor false but literally senseless.

This arbitrarily eliminates as meaningless any statements about the Reformed understanding of existence of God, which was the original topic of this thread. I realize that this is a separate criticism from the earlier one that verificationism is self-refuting, but I feel it is still relevant to demonstrating how verificationism begs the question against theism.

[Furrowed Brow]

Given these definitions, the verification principle does not meet the definition of an analytic proposition, and is therefore self-stultifying given my earlier comments.

Lets take the Ayer quote you provided.

To test whether a sentence expresses a genuine empirical hypothesis, I adopt a modified verification principle . . .I require of an empirical hypothesis not that it should be conclusively verifiable, but that some possible sense experience should be relevant to the determination of its truth or falsehood. If a putative proposition fails to satisfy this principle, and is not a tautology, then I hold that it is metaphysical and that it is neither true nor false but literally senseless.

• A: If there is no possible sense-experience relevant to the determination of the truth or falsity of a statement then it is an empirical hypothesis.

I dont think it can be much clearer that A is a contradiction and that this meets both Copis definition and the Stanford definition.

• B: if a putative proposition has no possible sense experience relevant to determining its truth and falsity, and it is not a tautology, then it is still true or false.

If B forms a contradiction which it is, it meets Copis definition, and if you can see how it forms a contradiction then it meets the Stanford definition.

I feel you equivocated on the meaning of the word "empirical" as it concerns Ayer's quote. In the first sense, "empirical hypothesis" seems to suggest any non-analytical, non-tautologous factual proposition, for example, the statement "God exists."

Well God exists would be a putative proposition that fails to be an empirical hypothesis and would by Ayers lights be a metaphysical statement. But if I am equivocating (I hope not) I think Ayer is being very clear. An empirical hypothesis is a proposition about an empirical matter of fact. These propositions are probable but never certain, and as you say include any non-analytical, non tautological statement that is a proposition about matters of fact.

In the second sense, Ayer refers to empirical evidence that could be used to justify such a proposition. This is not tautologous.

Not quite as you put it I think. The sense"experience relevant to whether the statement is true or false justifies that the statement is an empirical hypothesis about facts of the matter. He is not saying there need be empirical evidence in support of a statement to make the statement a meaningful proposition, he is talking about what is needed so that a statement has the logical form of an empirical hypothesis. And yes it is tautologous.

This arbitrarily eliminates as meaningless any statements about the Reformed understanding of existence of God, which was the original topic of this thread. I realize that this is a separate criticism from the earlier one that verificationism is self-refuting, but I feel it is still relevant to demonstrating how verificationism begs the question against theism.

I dont think it is wilful or arbitrary on Ayers part, I think it is the logical consequence of his analysis of meaning and what he counts as senseless. True his analysis may be insufficient to grapple with every aspect of language. He only admits two kind of non senseless statements, analytical and empirical hypothesis. But given the terms he allows himself and the definition he gives those terms his principle counts. Verifcationism may beg some questions about how language works but it " as far as I judge " hits the mark against anyone attempting to use language in an analytical way in the aid of theism. If theists want to play by the rules of analytical philosophy they need an analytical theory of language to show how their statements are meaningful.

Personally I prefer Wittgenstein would admits his argument is nonsense and if you like self refuting. I think Wittgenstein went way deeper than Ayer. But as far as Ayer goes I suggest you give him the credit of having a first class mind and would not easily fall into simple trap that means he self refutes. If he is guilty of confusion it is not a superficial one. And yes the Reformed Understanding the OP describes is exactly the kind of theology I think Ayer had in mind and would think his principle shows to be nonsense. And I think he would be right.

[Haven]

To test whether a sentence expresses a genuine empirical hypothesis, I adopt a modified verification principle . . . I require of an empirical hypothesis not that it should be conclusively verifiable, but that some possible sense experience should be relevant to the determination of its truth or falsehood. If a putative proposition fails to satisfy this principle, and is not a tautology, then I hold that it is metaphysical and that it is neither true nor false but literally senseless.

Well God exists would be a putative proposition that fails to be an empirical hypothesis and would by Ayers lights be a metaphysical statement. But if I am equivocating (I hope not) I think Ayer is being very clear. An empirical hypothesis is a proposition about an empirical matter of fact. These propositions are probable but never certain, and as you say include any non-analytical, non tautological statement that is a proposition about matters of fact.

As you admit, Ayer held that there can be only three types of propositions:

1) Analytical propositions, which are self-evident and cannot be coherently denied

2) Empirical propositions, which are all non-analytic propositions that concern matters of fact.

3) Non-analytic propositions that do not concern matters of fact -- this category is not relevant for the purposes of this post.

Therefore, it seems that the verification principle as stated by Ayer does contain an equivocation, as "empirical" is used in the first sense (i.e., in the term "empirical proposition") to denote "all factual propositions that are not analytic," while "empirical" in the second sense (i.e., in the term "empirical evidence"), is used to denote sense data. I feel this is enough to defend my contention that verificationism is self-destructive as it is not tautological, and therefore, not an analytical statement. However, I feel more arguments can be advanced against the verification principle.

Consider your statement:

A: If there is no possible sense-experience relevant to the determination of the truth or falsity of a statement then it is an empirical hypothesis. I dont think it can be much clearer that A is a contradiction and that this meets both Copis definition and the Stanford definition.

Given Ayer's definition of "empirical hypothesis," I don't think this follows. Consider the following examples:

a) The proposition "Haven is wrong, the earth is not 6,000 years old" meets Ayer's definition of an empirical proposition, because it is a non-analytic sentence. A reasonable person would find it necessary to produce sense data to back up that statement.

b) Consider the proposition "the external world exists." The statement is "empirical," because it is not analytical, however there cannot be any sense data to back it up, even in principle, because all sense data ostensibly results from external reality. However, most individuals find it reasonable to conclude -- a priori -- that the external world does exist, even in the absence of the possibility for sense data to support such a proposition. This seems to counter the contention that the statement "if there is no possible sense-experience relevant to the determination of the truth or falsity of a statement then it is an empirical hypothesis" is contradictory. Additionally, on verificationism, such a statement would be "senseless," but it seems absurd to consider the proposition that the external world exist senseless.

B: if a putative proposition has no possible sense experience relevant to determining its truth and falsity, and it is not a tautology, then it is still true or false. If B forms a contradiction which it is, it meets Copis definition, and if you can see how it forms a contradiction then it meets the Stanford definition.

See my above comments on the "external world" example. It defeats "B" as well.

[theopoesis]

I've seen a few posts on the thread discuss determinism, but little has been said about probabilism. I think the former has already been sufficiently elaborated, but the latter is worth noting. If we are in a naturalistic society which operates according to statistical probability and not according to mathematical necessity, then all things would seem to be contingent. However, if something is contingent, it could have been otherwise and thus could not be necessary. Insofar as our thoughts are material outcomes of a probabilistic process, they are contingent. If our thoughts about "necessary truths" are nothing but contingent probabilistic outcomes or material processes, then they cannot be "necessary." This seems to me a decently large epistemological problem for probabilistic materialism/naturalism.

[LiamOS]

I've seen a few posts on the thread discuss determinism, but little has been said about probabilism. I think the former has already been sufficiently elaborated, but the latter is worth noting. If we are in a naturalistic society which operates according to statistical probability and not according to mathematical necessity, then all things would seem to be contingent. However, if something is contingent, it could have been otherwise and thus could not be necessary. Insofar as our thoughts are material outcomes of a probabilistic process, they are contingent. If our thoughts about "necessary truths" are nothing but contingent probabilistic outcomes or material processes, then they cannot be "necessary." This seems to me a decently large epistemological problem for probabilistic materialism/naturalism.

It only becomes a problem if one attempts to terminate the probabilistic process at some point. Arguably not doing so is also a problem, but I'll give that some more thought later.

I'd also definitely like you to elaborate on this if you could; I've yet to see any significant discussion on this at all, and your thoughts are usually well presented and reasoned.

[Furrowed Brow]

As you admit, Ayer held that there can be only three types of propositions:

1) Analytical propositions, which are self-evident and cannot be coherently denied

2) Empirical propositions, which are all non-analytic propositions that concern matters of fact.

3) Non-analytic propositions that do not concern matters of fact -- this category is not relevant for the purposes of this post.

Sorry my bad. Ayer has no room for 3 I think.

Therefore, it seems that the verification principle as stated by Ayer does contain an equivocation, as "empirical" is used in the first sense (i.e., in the term "empirical proposition") to denote "all factual propositions that are not analytic," while "empirical" in the second sense (i.e., in the term "empirical evidence"), is used to denote sense data.

Well you can now probably shoot me out of the water but I dont think Ayer uses the phrase empirical evidence. There is sense-experience and there are empirical hypotheses and an empirical hypothesis is a statement which has the possibility of a relevant sense-experience. He is being perfectly systematic as far as I can tell.

I feel this is enough to defend my contention that verificationism is self-destructive as it is not tautological, and therefore, not an analytical statement.

Well frankly no. You have not shown what the equivocation is and how it is dangerous to the verication project. You need to take Ayers actual formulation, and show how it is not tautological.

And again I think you are just underestimating the calibre of the fellah if you think he is quilty of equivocation.

Also I note the goal posts have now drifted a little. Previously you were arguing that on its own terms verification needed verification, I think you are now recognising Ayer was attempting an analytical principle, and if successful verification does not apply to itself.

a) The proposition "Haven is wrong, the earth is not 6,000 years old" meets Ayer's definition of an empirical proposition, because it is a non-analytic sentence. A reasonable person would find it necessary to produce sense data to back up that statement.

Yes. But to meet Ayers there only need be principle some possible relevant sense experience for the statement Haven is wrong, the earth is not 6,000 years old to be an empirical hypothesis. So yes to be actually evidenced either way would make the question tangible and not abstract, but all there need be to be an hypothesis is that it is in principle possible to find relevant sense experience, which in the case of the geological age of Earth translates as there is some possible relevant data.

b) Consider the proposition "the external world exists." The statement is "empirical," because it is not analytical,

Being not analytical does not make a statement empirical. What makes a statement an empirical hypothesis is that there be some possible sense-experience statement relevant to deciding whether the statement is true or false. I think Ayer would find this statement "the external world exists" either senseless or redundant depending on what the author hopes to invoke.

however there cannot be any sense data to back it up, even in principle, because all sense data ostensibly results from external reality.

And it is exactly this kind or assumption that the use of the word external leads us in to that I was warning against. You are free to make that assumption but I think Ayers verification is not predicated on some implicit theory of an external reality. This is to misunderstand Ayer.

However, most individuals find it reasonable to conclude -- a priori -- that the external world does exist, even in the absence of the possibility for sense data to support such a proposition. This seems to counter the contention that the statement "if there is no possible sense-experience relevant to the determination of the truth or falsity of a statement then it is an empirical hypothesis" is contradictory.

Sense data is bumpy and you can have the experience of stubbing your toe. When the statement is of the kind FB stubbed his toe on a rock this is an empirical hypothesis to which there is some possible sense experience relevant to the claim. If however the statement is FBs experience of stubbing his toe is due to an external reality there is either redundancy and this means exactly what the first sentences means, or if it is trying to invoke some ontological distinction this is exactly the kind of thing verification should be a bulwark against because to talk about a reality external to the sense experience requires metaphysics. And as you know Ayer has no truck with metaphysics.

This does not mean there is anything wrong with the verificationism, it does mean verifcationism is showing us where the limits of language are to be found. It does not matter what we want to say, it matters what we can say sensibly. Ayer thinks metaphysicians fail to appreciate when they stop making sense.

Additionally, on verificationism, such a statement would be "senseless," but it seems absurd to consider the proposition that the external world exist senseless.

"FB stubbing his toe on a rock" is meaningful and makes sense, but "FB stubbing his toe on an external reality" is not meaningful and makes no sense at all if this is supposed to mean something not expressed by FB stubbed his toe on a rock . It is exactly this kind of logical mistake that Ayer is warning against.

see my above comments on the "external world" example. It defeats "B" as well.

It kind of indicates to me you have not really got to grips with what Ayer is about and what it means to accept metaphysics really is senseless. The kind of external word distinction you are invoking just becomes senseless if close attention is paid to how statements are made meaningful. So no you have not defeated Ayer, you have however shown you do not have a taste for what is implied. I think the kind of attitude you are demonstrating is common and a large reason for the drifting away from verificationism. A lot of folk just did not like what it meant, and the way it hampered what they thought they would like to say, so they just ignored it. And so much needless nonsense ensued.

[theopoesis]

Hi Aki:

I'll gladly try to clarify, if you'll do the same. I'm finding that I don't grasp your response.

It only becomes a problem if one attempts to terminate the probabilistic process at some point. Arguably not doing so is also a problem, but I'll give that some more thought later.

Yep, this part right here is what I don't understand. Little help?

I'd also definitely like you to elaborate on this if you could; I've yet to see any significant discussion on this at all, and your thoughts are usually well presented and reasoned.

Yeah as I re-read it my summary isn't the best. In fact, the way I worded things I'm pretty sure will not stand logically. It's been a long week. Let me try again...

For starters, I'm ignoring the possibility of free will, which is just too challenging to validate, theistically or naturalistically.

Moreover, I think I'm using "probabilism" in a non-technical sense. I may have made up its use in reference to what I'm trying to say. I guess, though, I'm pointing to the fact that modern science seems to have two different options. The first is deterministic, which from my limited knowledge seems to be more mechanical. Given a set of specific factors, a specific outcome is guaranteed. The thread seems to have been asking, given presupposing a set of exclusively naturalistic factors, can the determined of our brains as part of this deterministic process uncover necessary truth? That is an interesting question indeed. If our choices are not freely made according to what we consider rational, but are determined by a naturalistic determined causal chain, can we explain this causal chain in such a way that it validates the determined conclusions of our thought process?

However, I think there is a second variety of science, more in line with atomic theory, that looks at things through probability. Given a set of specific factors, a range of specific outcomes is possible along some distribution (standard or otherwise). The "laws of nature" are no longer guaranteeing a specific determinate outcome. So let's say that every thought that we have, given specific naturalistic factors, is the outcome of probability. It could have been otherwise. It was contingent.

The problem here is twofold. First, can naturalism explain a mechanism by which a thought process can achieve valid results? Of special significance would be the question as to whether these valid results would include metaphysical, speculative, philosophical thought or just practical thought (practical in the sense of concretely advancing survival)? This is the same challenge that determinism faces.

The second problem is what is unique to probabilistic thought. According to probabilism, it would seem that everything we think is a result of natural probability, one outcome among many. A thought chain could have just as easily been a different thought chain. Any given chain of thoughts, i.e. reasoning, could have been otherwise, and is therefore contingent. The problem is then more intense than for deterministic naturalism. Not only do we need to explain why a causally natural thought chain arrives at truth, we have to explain how this causal chain's conclusions are valid if they probabilistically could have been otherwise. If each and every conclusion we come up with is not necessarily caused, but one possible outcome of many, why is this conclusion any more valid than the other conclusion our mind could have naturally arrived at through another random possible outcome.

Idk if that makes any sense or not. I'm getting a headache (worked nights in addition to the normal 7-5 this week) so I think I'm going to tap out for the next few days and sleep. I'll check back when I return.

[sayak83]

As soon as free will is shown to make sense within the scope of our physical and biological understandings, I cannot take it seriously as a concept. My stance on this is much more adamant than most, as I view it as a common 'cop-out' on behalf of many Theists from explaining contradictions in their views.
I've yet to even have a Theist try to reconcile free will and physics, which is a little disheartening.

Determinism is self-refuting:

1) If one's thoughts are determined by prior blind causal chains beyond one's control, then one's reasoning process is also determined by such causal chains.

2) Any beliefs one forms on the basis of reason will be the result of such deterministic forces, as reason entails thought.

3) Deterministic physical forces have no end or goal, but are simply combinations of chance and necessity.

4) Chance and necessity are purposeless, while reaching truth is a purpose.

5) Therefore, there is no reason for one to believe that any of her beliefs -- including her belief in determinism -- is true.

Even if determinism were true, it would be irrational for one to accept it.

Therefore Google's search algorithm simply cannot output the most relevant search results for the words you type because that deterministic algorithm simply cannot aim at finding the true solutions of your search query. Yet it does.
Refuted.
Your argument is therefore refuted.

[AquinasD]

Therefore Google's search algorithm simply cannot output the most relevant search results for the words you type because that deterministic algorithm simply cannot aim at finding the true solutions of your search query. Yet it does.
Refuted.
Your argument is therefore refuted.

That's assuming Google's search algorithm is a kind of reasoning.