Carrier’s Theory of Knowledge

by Luke Muehlhauser on April 2, 2009 in Reviews

I’m blogging my way through Sense and Goodness Without God, Richard Carrier’s handy worldview-in-a-box for atheists. (See the post index for all sections.) Last time, I discussed section II.1 Philosophy: What It Is and Why You Should Care. Today, I discuss the introduction to section II.2 Understanding the Meaning in What We Think and Say.

I’m sure Carrier lost some readers by opening his book with a fairly technical discussion of the meanings of words and how they work. But this is where he must start. For if we are going to discuss how we can know things, and then what we know, we must carry on these discussions with words. So, we must understand what we are saying before we can say anything clearly at all.

A truth claim is called a proposition. For example: “It is raining” or “2+2=5″ or “Extraterrestrial life exists.” These are all propositions. “Stand up!” or “What’s your name?” are not propositions; they cannot be true or false.

Propositions make predictions

But Carrier defines propositions more precisely than this. He also says that (1) a proposition must have meaning, and (2) it must make predictions. That is, the proposition “I have a cat in my house” predicts that we should find a particular arrangement of bone and meat and fur in my house if the proposition is true.

This is crucial to Carrier’s philosophy:

Since understanding what words and sentences mean requires a mastery of language… and since understanding the predictions entailed by any given sentence requires a mastery of logic, these two studies must be fundamental to our education.

…if we can find any proposition that has meaning but does not make any predictions, or that makes predictions but does not have any meaning, or that can be confirmed as true or false without any reference to what it predicts, then this principle would have to be revised, and my entire philosophy reconstructed from the ground up…

Right away we see a problem that could demolish Carrier’s entire philosophy.

How do the axioms of math or logic make predictions? They don’t! Does this mean they are not propositions? Doesn’t this undermine Carrier’s philosophy?

On page 5, Carrier asks for “intellectual charity.” That is, if he appears to contradict himself, then “whatever interpretation would eliminate the contradiction and produce agreement is probably correct.” If there really is a mistake, then we are encouraged to contact him.

So, I kept reading to see if Carrier would make sense of this apparent problem – that the axioms of math or logic don’t make predictions, and thus aren’t propositions under Carrier’s philosophy.

I found the answers scattered throughout the chapter, and I’ll collect them here:

  • Page 28: “[Predictions include] the experience of ‘equivalence’ or ‘tautology’, which is the experience that ‘this’ is identical to ‘that’… and [also] historical statements, since an explanation of something that has already happened is just a prediction projected back in time.
  • On pages 35-37, Carrier specifies that a rhetorical tautology like “Cats are felines” predicts that in any case where we use the word “cats,” we could use the word “felines” to refer to the same thing (in American English, anyway).
  • On pages 42-43, Carrier says that contradictions predict something we will never and can never experience, for example: “this cat is a herbivore and a carnivore.” We know this because, from the rules of language, we know that the above statement really means “this cat is a herbivore and a not-herbivore,” something that cannot be experienced. So, the statement is false. Apparently, contradictions predict something about our language, which we can experience as false.
  • Page 53: “…when a proposition of logic or mathematics is challenged and seriously debated, the most widespread and solid agreement is achieved in comparison with any other method or subject. This is because the predictions entailed by such propositions are comparatively few, simple, and precisely defined, as well as thoroughly interrelated, and therefore these propositions are very easy to test… propositions of logic and mathematics only make claims about the meaning of concepts. So the only empirical inquiry they require is conceptual, and therefore inexpensive and immediate. For they can all be tested in the laboratory of the mind, where concepts exist.”

Here, Carrier seems to use uncommon definitions for words like “prediction” and “empirical” that include conceptual manipulations. By changing their meaning he remains consistent, but it is confusing for him to do so. Who else would call mathematical calculations “empirical inquiry”? Usually, “empirical inquiry” means that we look and see what is “out there” in the “real world.”

But maybe it’s not so odd. Mental experiences are “out there in the real world.” They really exist. Consider his account of how different types of propositions make predictions, on page 41:

All the other types of proposition can actually be described as types of hypothesis, as different kinds of claims to fact. For they all make predictions. References (lexical definitions) predict that if any copy of the implied lexicon is consulted… the stated meaning will be experienced. So “cats are felines” is true if the code word ‘cat’ is actually found to be equivalent to the code word ‘feline’ in some actual lexicon…

Descriptions… predict that if the thing being described is experienced, the features stated in the description will also be experienced. [If "Roses are red," then wherever roses are experienced, redness should also be experienced. Of course this is false, as some roses are not red.]1

In the case of opinions, they claim that someone actually has the stated opinion.

But if propositions must make predictions, what about the axioms of logic or math? What about people who have mental experiences that are illogical? How do logic and math retain their purity if they are reduced to “experiences” in the mind, which may vary from person to person?

It seems Carrier would say that the axioms of logic predict experiences of coherence or possibility in our minds. For example, the principle of contradiction predicts we will not be able to imagine the existence of something that is both A and not A at the same time in exactly the same way. But what if someone is illogical, or uses Zen logic? I suppose Carrier would reply that the axioms of logic make predictions that come out correct if our cognitive faculties are working correctly. But doesn’t this beg the question? How are we to say our thinking is “correct” before we demonstrate that logic is correct?

Carrier seems heavily influenced by logical positivism, which diminished in the 1960s due to at least 5 problems:

How does Carrier answer these objections? And why does he assert his view as superior to the many competing views? It’s not surprising that he ignores them in a book written for the layman, but in general I am skeptical of his method for gaining knowledge, and therefore of his entire project.

Carrier’s book is subtitled “A Defense of Metaphysical Naturalism.” Because his book does not take the time to answer common objections, it is only barely a “defense” of anything. It is more like a presentation of his worldview, which is also very valuable. A thorough defense of everything covered in this book would span thousands upon thousands of pages!

So, understanding (but not necessarily accepting!) what Carrier means by “prediction” and “experience,” we are ready to move on to the next topic: the meaning of words.

  1. I’d like to clarify for Carrier that “Roses are red” may have to be taken to mean something like “Rose petals are red when they are biologically normal and white light is shining on them.” []

Previous post:

Next post:

{ 22 comments… read them below or add one }

Jack April 3, 2009 at 4:42 am

It's my understanding that anything properly understood as an axiom is something that cannot be false. An axiom's truth is simply posited in a given framework. *Other* propositions are built on the axioms, make predictions, and can properly be falsified, but the axioms can't be. So, it seems that there's a sense in which they actually *don't* make predictions – they more or less fully determine the predictions that the non-axiomatic propositions make. If the definition of "proposition" requires that propositions make predictions (does it really?), then it would appear that axioms don't qualify.

Maybe? Just exploring the objection :).

  (Quote)

mikespeir April 3, 2009 at 6:54 pm

I have and read the book about a year ago. I'll admit that I'm not really sophisticated enough in this kind of thing to give it a proper analysis, but I do recall feeling a bit uncomfortable about some things Carrier said. I'll be interested in the rest of what you have to say about it.

  (Quote)

Ryan April 3, 2009 at 10:29 pm

I don't necessarily agree with verificationists (that all statements which are not empircally verifiable or true by definition are "meaningless").

However, I think that we can only make a claim to truth when something is logically true or empirically verifiable. This is because the laws of logic (such as A=A) are self-verifying. Their truth is independent of anything else.

Experience is likewise self-justifying, since you cannot deny your own experience (or memories).

  (Quote)

Ben April 4, 2009 at 8:15 am

"I suppose Carrier would reply that the axioms of logic make predictions that come out correct if our cognitive faculties are working correctly. But doesn’t this beg the question? How are we to say our thinking is “correct” before we demonstrate that logic is correct?"

I think he addresses this kind of issue with his talk on Cartesian demons. The point is you can't do better than what you can possibly do. And knowing that you know you are right independently of your own fallible knowledge base is clearly impossible. Just like on your post on morality detectors, you frame your inquiry in a way that cannot possibly find an answer that can even in principle ever be true.

"I do not consider it a “defense” of anything"

Why do you have to be so extreme about so many things? *sigh* Even if it were true, you could have said it's at least a *partial* defense of metaphysical naturalism. Your absolutist rhetoric would indicate that he doesn't defend anything in his presentation. Clearly that's not the case and yet we wouldn't know that from what you've said. What do you have against nuance? What did it ever do to you? :p

Ben

  (Quote)

lukeprog April 4, 2009 at 3:00 pm

You may be right about how Carrier addresses the axioms of logic via experience. Hopefully Carrier will elucidate his position in future writings.

I think you\\'re right about \\"defense\\", so I changed the wording in my post.

As you can see from the above, Ben, I do appreciate your comments. But I\\'m not against nuance, nor am I \\"extreme\\" about everything. For example, I\\'m not as extreme as Carrier about the historicity of Jesus or the theology of Paul. Another example is that after leaving Christianity I had no idea what to think about the morality of abortion – arguments on both sides were really bad. If desire utilitarianism is true, it would turn out that I\\'m mostly pro-life, though not as strongly as my Christian parents. I came to similar positions through two perspectives and methods that couldn\\'t be more different (Biblical Christianity vs. deterministic physicalism).

Granted, a lot of nuance must be sacrificed to keep these posts short, but I will try to keep the most important nuances in, as I\\'ve done by editing the above post in response to your comment.

Also, I\\'ll have another try at arguing against feelings in moral epistemology, and I look forward to seeing how you respond to that.

  (Quote)

Ben April 5, 2009 at 9:37 pm

I said:  “Why do you have to be so extreme about so many things?”

You said:  “But I’m not against nuance, nor am I ‘extreme’ about everything.”

:p  I forgive you. 

There’s no need to take jabs at Carrier especially since he understands the burden of proof is on his shoulders and he fully expects to have to actually take the argument through the appropriate channels of peer review.  There’s a difference between holding an extreme position and holding it for extreme reasons while being unable to appreciate relevant nuances that can throw off the sensibility of an entire position.  For instance, if Carrier thought Jesus was a myth just because the early epistles have little to say about a historical figure, then that would be relevant.  Or if he was so dull as to think all of the gospels must be false because some parts are, that would also be relevant.  However, I think you’ll find his discussion on the subject matter quite sober and devoid of such flippancy (though you might never know that from listening to some of his critics, haha).  The link to my “website” is an audio clip of him on the Infidelguy show talking about how *not* to defend Mythicism, if you are at all curious (i.e. no obligation). 

I totally understand wanting to make sure you come to your own moral conclusions and wanting to avoid following just a different herd uncritically.  I had much the same attitude (to a fault in some cases) when I left Christianity behind.  Of course, there may be good reason a herd of reasonable people tends to think the way they do and it’s no crime to agree with a correct answer.  I’m sure you know that.  It’s not like you need my permission to take your time figuring things out for yourself and I’m glad my some of my comments have been helpful.

Ben

  (Quote)

Ben April 5, 2009 at 9:39 pm

Oopsie.  Wrong link.  Here.

  (Quote)

lukeprog April 5, 2009 at 10:21 pm

Thanks, I think I’ve heard that Infidel Guy bit before.

I’m only “extreme” where I think reason and evidence warrant it. For example, I’m an “extreme” advocate of cell theory (everyone is). If you want me to be less extreme about denying the role of moral feelings in gaining moral knowledge, you’ll have to give me some good reason to think that moral feelings are a good way to gain moral knowledge.

  (Quote)

Ben June 19, 2009 at 6:03 am

lukeprog: Thanks, I think I’ve heard that Infidel Guy bit before.I’m only “extreme” where I think reason and evidence warrant it. For example, I’m an “extreme” advocate of cell theory (everyone is). If you want me to be less extreme about denying the role of moral feelings in gaining moral knowledge, you’ll have to give me some good reason to think that moral feelings are a good way to gain moral knowledge.

Okay, that’s easily done.  Since desire utilitarianism is about properly managing an economy of desires, it naturally follows that the general fuzzy experience of having gone through life and having some idea of what works and what doesn’t is at least a preliminary indicator of what may in fact be moral in a more refined and hammered down philosophically nuanced position.  It follows that our moral feelings have something in the ballpark at least of proper calibration.  Of course they can be riddled with stupid prejudices as well, but that’s certainly not the case for everyone and anyone has to start somewhere.
How do I track these comments on all your posts without having to create an RSS folder in google reader with just your comment threads?  That’s crazy.

  (Quote)

lukeprog June 19, 2009 at 7:06 pm

Ben,

Are you saying that we can to some degree trust our moral intuitions AFTER we have accepted the truth of desire utilitarianism? If so, that is quite than our base moral intuitions. That is more like making educated guesses about the distribution of desires, which is necessary for all of applied ethics until we have done better research under the model of desire utilitarianism.

  (Quote)

Ben June 21, 2009 at 10:13 pm

lukeprog: Ben,Are you saying that we can to some degree trust our moral intuitions AFTER we have accepted the truth of desire utilitarianism? If so, that is quite than our base moral intuitions. That is more like making educated guesses about the distribution of desires, which is necessary for all of applied ethics until we have done better research under the model of desire utilitarianism.

I’m using desire utilitarianism to shed retroactive light on what moral people presumably have been trying to do for thousands of years.  Wouldn’t you expect a good theory to be able to do that?
Ben

  (Quote)

lukeprog June 22, 2009 at 7:30 am

Ben,

I don’t think so, since I think people have been systematically wrong about probably MOST moral judgments. But I think I may be misunderstanding you altogether.

  (Quote)

Ben June 22, 2009 at 1:08 pm

lukeprog: Ben,I don’t think so, since I think people have been systematically wrong about probably MOST moral judgments. But I think I may be misunderstanding you altogether.

I don’t understand the perspective, “Some people get things way wrong, therefore totally ignore any previous background experience in life when assessing moral questions.”  I give people partial and half assed credit in whatever abstract way is applicable to whatever extent is justified.  That’s finding common ground and I don’t focus so much on how “wrong” they are as much as I note their moral inefficiency (in relation to what I currently think the most objective moral theory is).   I’m thankful for the tool of “desire utilitarianism” to do this (in addition to Richard Carrier’s “goal theory”) and only have this one significant beef so far with Alonzo and yourself that is a bit frustrating.  It’s really hard to be completely wrong about absolutely everything and anyone that thinks they are completely getting away from the mistakes inherent in their misapprehended background knowledge is deceiving themselves.  We have to move forward based on something and that’s always in light of our previous experiences in one way or another.  The human brain just doesn’t work some other way.  No one truly starts over.  We make transitions.  When you deconverted from Chrisitianity you claimed to think you knew nothing of morality and I find this to be completely implausible.  If we compared your beliefs then to your beliefs now, I’m certain I could find wide areas of agreement in various ways and could easily construe many Christian moral beliefs in light of how well they managed a flourishing economy of desires.  As I understand it, you then asked yourself a question concerning whether you had inadvertantly done more harm than good up to that point as though your previous background knowledge in morality had absolutely *nothing* to do with the formulation of that question.  You seem to be arguing for things that just aren’t humanly possible and vastly overstating your perspective for some kind of unevaluated arational reason, in my opinion.  I’ll understand if you are still a little skeptical, but part of moral wisdom, as I see it is learning not to create extra problems for the sake of the solutions.
thanks for listening anyway,
Ben
BTW, is there some kind of tutorial on your site on how to use html properly?  Nothing I seem to do works.

  (Quote)

lukeprog June 22, 2009 at 3:51 pm

Ben,

Do you see the buttons for bold, italic, links, etc. above the comment box? You can also hit the HTML button if you prefer to compose in HTML.

  (Quote)

Ben June 22, 2009 at 11:30 pm

Okay, I’m going to attempt a paragraph break…
 
 
How’d I do?

  (Quote)

Ben June 22, 2009 at 11:31 pm

K, that was overkill.
Why doesn’t it just work when I hit enter normally?  We’ll see if it does…

  (Quote)

Ben June 22, 2009 at 11:33 pm

Nope.  And hence my confusion.  How do you get normal paragraph breaks without inserting overkill html?  If I do nothing but hit enter, it gets mushed and if I attempt to correct it by inserting <br>’s and <p>’s it over-does it.  What am I missing?

  (Quote)

lukeprog June 23, 2009 at 7:09 am

Hmmm.

I hit enter twice when I’m writing a new paragraph.

I don’t manually type HTML markup at all.

  (Quote)

Ben June 23, 2009 at 5:25 pm

Oh!  So you double space to achieve a single space.  Let’s try that.
 
How’d I do?

  (Quote)

Ben June 23, 2009 at 5:25 pm

Yay!  Finally!

  (Quote)

Ben June 23, 2009 at 5:26 pm

Oh, I see.  The comment box is set to “double space” when you hit enter once but the comment posts in single space, so that’s why we have to compensate.  Is there some advantage to that I don’t understand?
 
 

  (Quote)

lukeprog June 23, 2009 at 6:18 pm

Ben,

Unfortunately, I’m not a web designer or programmer… I barely have this site patched together as it is. :)

  (Quote)

Leave a Comment