Neo of The Skeptic Arena read two of my posts: “Not All Good Arguments Are Logically Sound Part 1” and “Not all Good Arguments Are Logically Sound Part 2.” (At first he only read Part 2, which might have led to confusion.) We had an email conversation that he decided to post on his website for educational reasons. Go to his website to see for yourself. He concludes the following:
Against my better judgment I went ahead and read part 1. As I expected, it was pretty much the same drivel I put up with in Part 2. It was a waste of time. Since he is obviously unable to defend his assertions, continuing the email exchange would also be a waste of time.
I suspect that Neo does not understand what “logically sound” means in this context, and I had a conversation with someone else who had the same problem. Dictionary definitions are not appropriate for this conversation. I defined “logically sound” in Part 1. I will say more about that in what follows.
The conversation continues.
I will now respond to what he said last. Keep in mind that he said all of these things at once before I responded to him here. I will quote several things he says and reply to him below:
James, your reply ignored my question: how could any argument be “meant” to be unsound? And other than you, who made that claim?
My reply: Inductive arguments are not meant to be sound. That is true by definition. You learn that in a logic class. See “Deductive and Inductive Arguments” at the Internet Encyclopedia of Philosophy.
A sound argument is a logically valid argument with true premises. Logically valid arguments have an argument form that couldn’t possibly have true premises and a false conclusion at the same time. For example:
- All dogs are lizards.
- If all dogs are lizards, then all dogs are reptiles.
- Therefore, all dogs are reptiles.
The first premise is false and the conclusion is false. It doesn’t matter—it’s still a valid argument.
This is simply the definition of “valid” and “sound.” You learn this in a logic class. See “Validity and Soundness” at the Internet Encyclopedia of Philosophy.
James, good scientific theories are supported by evidence not arguments.
My reply: Yes, they are supported by arguments. Good scientific theories are conclusions—they should be believed because they are highly justified. Sufficiently good arguments provide us with conclusions that we should accept based on our justifications. An argument is the reasoning process made explicit. There is inductive reasoning and deductive reasoning. Inductive arguments and deductive arguments correspond to these two types of reasoning.
Let’s say a scientist predicts that gravity will cause objects to that are dropped fall at a certain speed at an accelerated rate (in a vacuum while at a certain distance from the center of the Earth). Scientists can then check that prediction using experiments. If objects always fall at that speed in experimental conditions, then the scientists can have the following argument:
- Dropped objects always fell at a certain speed at an accelerated rate in our experiments.
- The belief that dropped objects always fall at a certain speed at an accelerated rate is consistent with all observations we have.
- Therefore, probably all dropped objects will fall at a certain speed at an accelerated rate.
This is an inductive argument. Inductive arguments take observations and make predictions (or generalize) based on those observations. Inductive arguments can only tell us what’s probably true and there’s always a chance they have false conclusions.
For example, at one point many people thought that all swans were white because they only saw white swans. Even so, a black swan was eventually found. That proved the conclusion to be false.
In the first email James had written:
However, not all good deductive arguments are logically sound either.
To which I replied:
“If they aren’t, then I would be even more interested in exactly
how you are defining ‘good’.”
Now James responds:
A good reason to believe something, as stated earlier.
James, why would I have good reason to believe a deductive argument that wasn’t logically sound?
My reply: Because the premises of the deductive argument can be justified without necessarily being true. Sound arguments have to have true premises. For example, arguments can use the highly justified conclusions of science as premises.
Consider the following argument:
- Objects that are dropped fall at a certain speed at an accelerated rate.
- If objects that are dropped fall at a certain speed at an accelerated rate, then dropping my glasses will cause them to fall at a certain speed at an accelerated rate.
- Therefore, dropping my glasses will cause them to fall at a certain speed at an accelerated rate.
This is a deductive argument that uses a conclusion attained by science as a premise—that objects that are dropped fall at a certain speed at an accelerated rate. This premise is probably true, but it could be false. Perhaps no objects will fall in the future at all. That seems unlikely, but it’s perfectly reasonable to believe the glasses will fall as predicted.
Now from James’ first email where he had stated:
There is a sense that deductive arguments would ideally be logically sound, but some deductive arguments have sufficiently justified premises, even if those premises aren’t known to be true for certain.
To which I had replied:
“Only religious nut jobs believe in absolute certainty. Since you are
not one of them, why do you even bother mentioning a concept
that is unknowable (absolute certainty)?”
Now James responds:
Because it has to be said to prove what I want to prove.
My reply: To require that good arguments are sound is to require the premises of good arguments to be known for absolutely certain. Sound arguments have to have true premises. That’s true by definition. It’s what “sound” means in this context.
James, how does resorting to an imaginary concept prove what you want to prove?
[James replied: If we can't know things for certain, then we can't depend on all our arguments being sound.]
James, name one thing that you know “for certain.”
We can know things only to a degree of certainty, as your earlier statement demonstrated when you correctly stated that science is often proved wrong; yet we can depend on our arguments being sound if we rigorously follow the rules of logic.
So I disagree.
My reply: You are wrong. The rules of logic can’t guarantee that our premises are true, but sound arguments have to have true premises.
If you don’t rely on sound arguments to form your beliefs, what do you rely on?
[James replied: We rely on the best information available and we make predictions.]
James, why can’t that “best information available” be a sound argument?
[James replied: Because our best scientific conclusions are not necessarily sound but we still have a good reason to believe them.]
James, name one.
That will require that you produce a scientific conclusion that is either 1) invalid, or 2) contains a false premise.
My reply: A conclusion can’t be invalid or contain a false premise. You mean that the argument must be invalid or contain a false premises. Inductive arguments are technically invalid. The truth of the premises of inductive arguments does not guarantee that the conclusions are true. I already gave examples.
James now quotes me from email #1:
“When your argument has to resort to counterfactuals, you should
realize that your argument is in deep trouble.”
I disagree. The point is we don’t rely on it being sound.
James, why would you rely on something that wasn’t sound?
[James replied: We rely on inductive evidence instead.]
James, evidence is neither inductive nor deductive – reasoning is.
I found one web site where that term was used … incorrectly.
The author defined observations as inductive evidence showing that he doesn’t understand the meaning of induction.
My reply: Yes, I do understand the meaning. If you want to prove otherwise, go ahead. See “Deductive and Inductive Arguments” at the Internet Encyclopedia of Philosophy.
Update (9/9/2012): I made a few clarifications and rephrased some of what I had to say. Our responses were also mixed up a bit and had to be separated.
Update (9/12/2012): I explained that I am responding to something he wrote all at once.