If P and Q, then...what?
November 6, 2009 6:39 PM Subscribe
How do I get better at recognizing logical fallacies...
...particularly in conversation? In the midst of conversation I usually find myself responding to the premises or conclusions of an argument by countering with different premises or conclusions rather than noticing that the argument itself is wrong (or weak) to begin with. When I reflect on the conversation later, I can usually see how an argument was flawed, but while it's happening, forget it. This really annoys me.
Has anyone else overcome this problem? If so, how? Is there anything I can read or any exercises I can do to build up this mental muscle?
...particularly in conversation? In the midst of conversation I usually find myself responding to the premises or conclusions of an argument by countering with different premises or conclusions rather than noticing that the argument itself is wrong (or weak) to begin with. When I reflect on the conversation later, I can usually see how an argument was flawed, but while it's happening, forget it. This really annoys me.
Has anyone else overcome this problem? If so, how? Is there anything I can read or any exercises I can do to build up this mental muscle?
If Meadowlark's link is too long for you, you can just focus on Schopenhauers 38 strategems, or 38 ways to win an argument (none of them honest).
posted by alms at 7:38 PM on November 6, 2009 [3 favorites]
posted by alms at 7:38 PM on November 6, 2009 [3 favorites]
Find a political essay with which you disagree and then attempt to write an analysis of it which supports its conclusions.
Example: a die hard capitalist reads a chapter of Das Kapital and then argues in favor of Marx's ideas.
This will force you to (1) contend with the structure of the arguments the essayist writes, and (2) the structure of your own arguments.
posted by dfriedman at 7:48 PM on November 6, 2009
Example: a die hard capitalist reads a chapter of Das Kapital and then argues in favor of Marx's ideas.
This will force you to (1) contend with the structure of the arguments the essayist writes, and (2) the structure of your own arguments.
posted by dfriedman at 7:48 PM on November 6, 2009
From the post, it sounds like you understand the fallacies already, you're just not on your guard enough for them. Treat an argument like that as you would a conversation with a salesman -- always be on your guard, listening out.
The other key thing is to try and keep a mental overview of the entire argument in your head. This isn't how we usually speak to one another, so it's hard at first. Usually we just deal with the last thing a person said, and that allows them to get away with some hideous fallacies that would leap out in print. So build up their argument in your head; don't worry if you go quiet while you do it, they'll be flattered you're thinking. If you have trouble fitting bits together in your mind as a whole, that's a good sign aomething's amiss. Spot what it is, and you're half way there.
posted by bonaldi at 8:47 PM on November 6, 2009
The other key thing is to try and keep a mental overview of the entire argument in your head. This isn't how we usually speak to one another, so it's hard at first. Usually we just deal with the last thing a person said, and that allows them to get away with some hideous fallacies that would leap out in print. So build up their argument in your head; don't worry if you go quiet while you do it, they'll be flattered you're thinking. If you have trouble fitting bits together in your mind as a whole, that's a good sign aomething's amiss. Spot what it is, and you're half way there.
posted by bonaldi at 8:47 PM on November 6, 2009
If you understand the logical fallacies themselves but are having issues with spotting them in verbal debate, look up "weasel words." These qualifiers are good markers for bullshit.
posted by cj_ at 9:49 PM on November 6, 2009
posted by cj_ at 9:49 PM on November 6, 2009
Logical fallacies are a good taxonomy of how to be wrong, but they're not always practical to refer to in real life. You might find it more useful to identify common techniques people use to win arguments without being right.
You're looking for more intuitively identifiable behaviour that is commonly used to conceal fallacies, or particularly common instances of a fallacy. The category of fallacy is academic. Watch out for these (for example):
Wit without message. Easily identifiable when they reuse or rearrange your words. They are not Mark Twain.
"We need to identify our competencies"
"I think we just need a little less 'identification' and a little more competency."
If you're not paying attention, this sound as though the person has scored a point in the argument. In reality it's one step below a pun and shows that the person is more concerned with reacting to your words than what you're saying.
Ultimatums. Demanding you agree or disagree with some specific point, or denounce some exaggerated caricature (or Hitler). Redefining an argument into a very specific yes-no question.
"So you think babies deserve to die?"
It forces the other person to respond, and looks like an easy point to score by denying the absurdity. But it's not meant to be an argument-winner. It's for rebooting the conversation and wiping out any progress you were previously making towards any kind of point.
It becomes a questions and answers session that you were lured into with an easy-to-dismiss point, but then come the leading questions and the rhythm of the conversation has bee re-established as Q&A about what you believe. When you inevitably stumble, they win by default without having any argument of their own.
posted by Lorc at 1:18 AM on November 7, 2009 [3 favorites]
You're looking for more intuitively identifiable behaviour that is commonly used to conceal fallacies, or particularly common instances of a fallacy. The category of fallacy is academic. Watch out for these (for example):
Wit without message. Easily identifiable when they reuse or rearrange your words. They are not Mark Twain.
"We need to identify our competencies"
"I think we just need a little less 'identification' and a little more competency."
If you're not paying attention, this sound as though the person has scored a point in the argument. In reality it's one step below a pun and shows that the person is more concerned with reacting to your words than what you're saying.
Ultimatums. Demanding you agree or disagree with some specific point, or denounce some exaggerated caricature (or Hitler). Redefining an argument into a very specific yes-no question.
"So you think babies deserve to die?"
It forces the other person to respond, and looks like an easy point to score by denying the absurdity. But it's not meant to be an argument-winner. It's for rebooting the conversation and wiping out any progress you were previously making towards any kind of point.
It becomes a questions and answers session that you were lured into with an easy-to-dismiss point, but then come the leading questions and the rhythm of the conversation has bee re-established as Q&A about what you believe. When you inevitably stumble, they win by default without having any argument of their own.
posted by Lorc at 1:18 AM on November 7, 2009 [3 favorites]
Learn about first order logic. Take note especially of these concepts, in the introduction:
1) Many statements in everyday speech take forms like "All a's have quality x," "There exists an a with quality x," and so on. There are different ways to ascertain the truth value of such claims: if there exists an a without quality x, the statement "All a's have quality x" cannot be true.
Of course, that example is itself a logical assertion: If all a's have quality x, if any given object is an a, then it possesses quality x. If you can prove the premise is true (i.e. the object is an a) and the conclusion is false (the object does not possess quality x), then the statement itself is false. The argument, as it were, is flawed.
There are sets of rules, like the one I have (clumsily) outlined above, for statements which include OR, AND, NOT. They are often expressed in an easy to read format as "truth tables" and even rules for more complicated ones like "x is necessary for y," "x is necessary but not sufficient for y," etc.
Not incidentally, this is also exactly what lawyers do! If you want practice spotting logical fallacies in an argument, read some practice LSAT questions (woah.) Or you could opt for something a bit less serious like the logic puzzles on this site. This one jumped out at me...
posted by Truthiness at 1:56 AM on November 7, 2009
1) Many statements in everyday speech take forms like "All a's have quality x," "There exists an a with quality x," and so on. There are different ways to ascertain the truth value of such claims: if there exists an a without quality x, the statement "All a's have quality x" cannot be true.
Of course, that example is itself a logical assertion: If all a's have quality x, if any given object is an a, then it possesses quality x. If you can prove the premise is true (i.e. the object is an a) and the conclusion is false (the object does not possess quality x), then the statement itself is false. The argument, as it were, is flawed.
There are sets of rules, like the one I have (clumsily) outlined above, for statements which include OR, AND, NOT. They are often expressed in an easy to read format as "truth tables" and even rules for more complicated ones like "x is necessary for y," "x is necessary but not sufficient for y," etc.
Not incidentally, this is also exactly what lawyers do! If you want practice spotting logical fallacies in an argument, read some practice LSAT questions (woah.) Or you could opt for something a bit less serious like the logic puzzles on this site. This one jumped out at me...
posted by Truthiness at 1:56 AM on November 7, 2009
It sounds like you're not really listening. This is the problem with many forms of discourse today. People aren't listening to each other speak, they're just waiting (if even that) for their turn to talk. If you want to understand what people are saying, listen to them.
posted by headnsouth at 5:15 AM on November 7, 2009
posted by headnsouth at 5:15 AM on November 7, 2009
Has anyone else overcome this problem?
Of course. That's why there are classes offered by various institutions on formal logic,
posted by Neiltupper at 7:18 AM on November 7, 2009
Of course. That's why there are classes offered by various institutions on formal logic,
posted by Neiltupper at 7:18 AM on November 7, 2009
Excessive use of metaphors or rigidly insisting to only argue within the confines of the metaphor is generally a decent indicator that the argument is deeply flawed. The only way I've found to tear the argument out of the context of the metaphor is to make a ridiculous comparison between the metaphor and the actual argument. Metaphors are sometimes illustrative, but they rarely serve as an accurate simplified substitute for the argument at hand.
The other common logical fallacy that is easy to recognize is virtually anything that is presented as absolutely true or absolutely false. Black and white thinking. Virtually nothing in this world is not some shade of grey. Nothing works absolutely perfectly. Absolute evil is thankfully rare. And the list goes on. For instance, the question is not socialism vs capitalism, it is how much of each and in what aspects of our society. It's not was the war in Iraq right or wrong, but on balance was it more right than wrong. The world is complex, any argument that fails to recognized the inherit utter complexity of it all is likely false.
posted by whoaali at 11:34 PM on November 7, 2009
The other common logical fallacy that is easy to recognize is virtually anything that is presented as absolutely true or absolutely false. Black and white thinking. Virtually nothing in this world is not some shade of grey. Nothing works absolutely perfectly. Absolute evil is thankfully rare. And the list goes on. For instance, the question is not socialism vs capitalism, it is how much of each and in what aspects of our society. It's not was the war in Iraq right or wrong, but on balance was it more right than wrong. The world is complex, any argument that fails to recognized the inherit utter complexity of it all is likely false.
posted by whoaali at 11:34 PM on November 7, 2009
Response by poster: Everyone -- thanks very much. You've all given me a lot to think about and some great resources, too.
posted by Maisie at 11:03 AM on November 8, 2009
posted by Maisie at 11:03 AM on November 8, 2009
People do make logical fallacies when they talk but its only very rarely that I would stop someone mid-flow to point this out to them. Even then I never say, "hey, you just affirmed the consequent dumbass!"
The formal logic that philosophers use to explain the rules of inference to their students serve a specific job, which is to help you pass Logic 101 and understand why you passed it and (for example) Timmy didn't. They are not meant to directly help you argue with people in practice.
Arguments formed in natural language are by their nature imprecise and open to interpretation (unlike the language of formal logic). Learning the skills and techniques sof argument in natural language is a never-ending work because context and even language itself are in always in flux. But you've already noticed this, however your conclusion is wrong, you don't need to treat natural language as if were more like the language of formal logic.
Rather, having a clear understanding of the rules of logical inference is only one of many building blocks that will help you, over time, to become a better debater and to understand your own opinions more clearly.
posted by munchbunch at 1:57 AM on November 9, 2009
The formal logic that philosophers use to explain the rules of inference to their students serve a specific job, which is to help you pass Logic 101 and understand why you passed it and (for example) Timmy didn't. They are not meant to directly help you argue with people in practice.
Arguments formed in natural language are by their nature imprecise and open to interpretation (unlike the language of formal logic). Learning the skills and techniques sof argument in natural language is a never-ending work because context and even language itself are in always in flux. But you've already noticed this, however your conclusion is wrong, you don't need to treat natural language as if were more like the language of formal logic.
Rather, having a clear understanding of the rules of logical inference is only one of many building blocks that will help you, over time, to become a better debater and to understand your own opinions more clearly.
posted by munchbunch at 1:57 AM on November 9, 2009
This thread is closed to new comments.
posted by Inspector.Gadget at 6:51 PM on November 6, 2009