Variables in Arguments as a Source of Confusion
I was reading an argument happening in the comments of an article about Light Table switching to open source. The argument was about freedom in relation to software, and it went basically something like this:
People who use OSX are less free [than Linux users], because they don’t have the freedom to modify their OS source code.
No, they have the exact same freedom. People who use OSX and people who use Linux both have the freedom to modify the source code of Linux.
I’m not entirely sure, but this conversation reminded me immediately of arguing about a tree falling and making a sound when nobody’s around to hear.
The first persons statement uses a variable in the place that the second persons statement uses a constant.
X’s freedom is [partially] a function of [X’s OS].
vs
X’s freedom is [partially] a function of OS_List. (where OS_List is just a list of the OSs that he could in principle modify, regardless of if he wants to or is using any of those OSs)
(Obviously OS_List is a variable as well, but with respect to each person it’s relatively unchanging).
I’ve seen this crop up in various conversations before—one person arguing using a variable where another person is using a constant (if that’s the right way to describe it).
How does one diagnose the problem with this argument, if there is a problem? Is it a similar problem to the Tree in the Forest problem? Is there a standard rationalist way to dissolve the dispute so that both parties can leave not only agreeing, but also having a high probability of being correct when they leave?
It seems to be known under the name of the equal treatment fallacy in various blogs and articles, although none of them are from particularly respectable sources. Other examples are the right of homosexuals to marry a member of the opposite gender, the right of soviet citizens to criticize the president of the USA, and Anatole France’s famous statement that “in its majestic equality, the law forbids rich and poor alike to sleep under bridges, beg in the streets, and steal loaves of bread”.
This seems worth adding to a list somewhere or making a more elaborate article about. Anyone?
At least, the label “equal treatment fallacy” seems like it represents well enough most cases and, with those examples, evokes a clear picture. It doesn’t seem to refer to all “variable vs constant” issues following this pattern, but close enough.
Nice! I wanted to mention that one but you beat me to it :-)
It seems like the Linux user (and possibly the Soviet citizen example, but I’m not sure) is… in a broader category than the equal treatment fallacy, because homosexuality and poverty are things one can’t change (or, at least, that’s the assumption on which criticizing the equal treatment fallacy is based).
Although, I suppose my interpretation may have been different from the intended one- as I read it as “the OSX user has the freedom to switch to Linux and modify the source code of Linux”, i.e. both the Linux and OSX user has the choice of either OS. Obviously the freedom to modify Linux and keep using OSX would be the equal treatment fallacy.
I’m reminded of Douglas Hofstadter’s bit about “do this!”
If Alice touches her finger to her nose and says “Do this!”, should Bob interpret this as an instruction to touch Alice’s nose, or to touch Bob’s nose?
My usual approach to this kind of argument is to go meta… specifically, to ask what it is we are trying to achieve.
If it really is “freedom” expressed in generic abstract units, then sure, they have the same number of units of freedom, and that’s all we care about… but this seems to describe precisely nobody in the real world. If it’s something else for which freedom is a usually-convenient proxy, this is a good time to unpack the proxy and think about what we actually want.
In the specific case, I would likely ask “why is it valuable to a Linux user to be able to modify Linux source code? Do OSX users get the same value out of being able to modify Linux source code? Do we care about freedom even when it’s valueless? Do we care about value?”
Clearly different people value different freedoms differently, so this has to be taken into account.
A bit of a (fake) formalization: Total freedom of a person_i = sum over j (freedom_j * usefulness of freedom_j to the person_i).
This takes care of most “gays can marry straight” arguments.
Fear me, for I applied linear algebra to ethics! Fi=U{ij}f_j
Oh, oh, and this can be extended to estimate freedoms of whole groups of people, only now we sum over all people and weigh each person’s freedom by their usefulness/importance to the society. It’s a dot product!
Oh, even more, let’s maximize the total freedom, given some reasonable constraints, which we assume to be linear, Voila, simplex method! Look what we get for free: the optimum is almost always a vertex, something like a utility monster or a group of them. Damn, and it started so promising! Oh well, back to the drawing board on that last part.
Normalize the results—put each person on a 0-1 scale, where 0 is them being unable to do anything at all that they want to, and 1 is them being able to do everything they want to. Don’t bother with dot products, just take the mean(probably the geometric mean, to preserve minority rights).
Sure, each element of the usefulness matrix is between 0 and 1. You need the freedom vector to quantify how valuable each freedom is in general before scaling it by the usefulness of it to each individual. Feel free to explicitly write out your expression for the overall freedom and try it out on some simple examples.
Given input of a list of the value of various freedoms, divide by the sum to normalize the result. I don’t care if they make each individual freedom worth 3 points or 3450121 points, but I hold it self-evident that all men are created equal, and normalize their values in the algorithm as a result.
Fun politics variant:
“Disallowing gay marriage does not limit freedom—a gay man can marry a woman and a lesbian woman can marry a man, just like straight people.”
Relevant Language Log post
“Same” means “different”.
When someone calls two things “the same”, the real claim is that the two things (which are necessarily different, else they would be one thing) both fall into some category. When they are called “different”, a line is being drawn between them, instead of around both.
There are many categories you can use to unite or split any collection of things in any way you like. Which of these categories matter and which are irrelevant doodles on the map cannot be answered merely by pointing to their existence.
The main flaw in both arguments is the implication that “freedom” is a scalar and is terminally valuable to everyone. There’s lots of dimensionality to one’s ability to choose various options, which prevents a meaningful comparison.
There’s no disagreement over what each is allowed to do, the disagreement is over the value to each of those choices. And it’s not much disagreement.
The OS X user has the same freedoms as the Linux user: both can modify Linux but can’t modify OS X. However, the ability to modify Linux is more useful to the Linux user than to the OS X user, and the inability to modify OS X is more of a nuisance for the OS X user than for the Linux user.
I think this is much less a moral issue than the parallel cases mentioned elsewhere in the discussion, because no one owes you an OS you can modify in the same way as you’re arguably owed the right to marry a partner you can love, the right to criticize the government that has power over you, or the right to find food to eat and a place to sleep, and because you can switch OS much more readily than sexual orientation, government, or level of wealth. It’s for that reason that those instances are well framed as instances of inequality or even inequity in how different people are treated, whereas the differences between OS X and Linux are differences between those OSes rather than between particular people.
To whatever extent people are unable to use a different OS, those moral issues reappear and a framing in terms of inequality between people becomes more relevant. If everyone were assigned an OS at birth and forcibly prevented from using another, then this would be just like those other cases.
In that model means that man and woman have different freedoms with regards to going to the toilet.
I think it’s much more practical to say that both gender can go to the toilet that corresponds to their gender than to say that men have the freedom to go to the men’s toilet and woman don’t have that freedom. Few woman feel discriminated for not being allowed in the men’s toilet.
I think I might describe this more as equivocation between freedom as scope of local action and freedom as degree of global constraint: we’re used to dealing with the word in terms of the categories of action that governments or cultural forces exclude from us, de facto or de jure, and not in terms of what the local environment allows us to do. (Prohibitions on shouting ‘fire’ in a crowded theater are rarely held to infringe free speech, to give a famous negative example.).
I’d hazard a guess that this is partly due to the common use of “freedom” as a political buzzword, almost always in the constraint sense.
I used to hear something similar in debates over gay marriage:
Gay person: “I only want to have the same right as a straight person: the right to marry the person I love.”
Gay marriage opponent: “No no, you already have the same right as a straight person: the right to marry a person of the opposite sex. If you also want the right to marry a person of the same sex, you’re asking for extra rights, special privileges just because you’re gay. And that simply wouldn’t be fair.”
Edit: bramflakes beat me to it.
Gay person: “No no, I want everyone to have the right to marry a person of the same sex, even straight people! Equal rights for all — that’s perfectly fair.”
Generally: “Other people are also allowed to live in a society optimized for my utility function, so where is the problem?”
The lesson for you: Write shorter.
1) Write shorter to be the first one.
2) Edit your answer later to include everything relevant, fix spelling errors, etc.
3) Congratulations, LW becomes StackExchange!
Okok. It was meant humorously.
But I don’t think that the conclusion of your insightful comment (which really could have been a Discussion post). is inevitable.
There still is evolution going on on the organizational level. Those organizational structures which provide long time benefit i.e. are efficient and sufficiently stable should win over structures that collapse due to the iron law.
Thus lets hope that LW manages to provide benefit and be stable. I wonder whether EY which after all started out as charismatic leader (in the socioeconomic sense) has a plan to transition LW toward that. But then he probably hopes that FAI will solve these kinds of problems for us.
This is closely related to a failure mode of communication discussed here recently. The error made by the “fascists” and “rakes” in the linked post is a special case of the OP’s error of assumed variables: in this case the hidden variable is whose morality applies.
My guess is that these miscommunications often arise from inadequate empathy or theory of mind. It’s very common for assertions to have hidden or assumed variables that are reflexive, that is, they refer to the speaker. Some people have the ability to automatically transfer the hidden referent when changing such an assertion’s subject, and others must to reminded to do so. (And indeed, in some cases some will argue that it’s not appropriate to transfer the referent!)
The second argument is technically correct but misses the implicit point of the first. If the first said, “It is best to use software which is free”, perhaps with some further justification, then the second argument would not address it.
I don’t think that’s a form of confusion that people can have without some serious mind-killing going on. If you can’t figure out how to stop being mind-killed, no amount of arguing is going to help. If you can, getting around this problem is trivial. Either way, there’s no point in bringing it to attention.