Friday, November 19, 2010

Possibility and the burden of proof

If I didn't run into this sort of arguments often enough, I wouldn't be writing about this. Some time ago, Gualtiero Piccinini (whose computability-related stuff I really enjoy reading) posted a question about known good arguments for the existence of afterlife. Joshua Carl Davis commented:
I make an argument in my book Metaphysics and the Meaning of Life that there is an afterlife. The argument goes as follows: 1) It is always valid to argue from actuality to possibility (a logical principle dating to Medieval times), 2) I exist now therefore it is possible for me to exist again. While this does not prove the immortality of the soul, it switches the burden of proof to the other side and, I believe it is an overwhelming burden. The argument is given on page 282. The book is available on Amazon. Good luck to you.
Eric Thomson gives a pretty straightforward explanation of what went wrong there.
  1. ET argues that equally well the fact that one didn't exist before they were born shifts the burden to the other side.
  2. ET points out that establishing possibility doesn't automatically shift the burden of the proof.
  3. ET indicates that the possibility of future existence doesn't follow from the actual existence by the principle that JCD refers to.
As for point 1, JCD responds:
As for your first point, which to you is more certain: "I exist now" or "I did not exist before I was born"? That should indicate which is the stronger argument.
The problem is, a reductio argument of this sort cannot be refuted by saying one of the premises is less certain than other. Suppose someone gives an argument: "P - therefore, by my magical strategy, Q", and someone else responds: "The magical strategy doesn't fly: I can equally well argue: Z - therefore, by the same magical strategy, not Q". Now, saying that Z is less certain than P doesn't help. As long as Z and P are both true, the reductio works and shows that the magical strategy (let's not hesitate to use the technical term) sucks.

What has to be done to defend the magical strategy? Well, once one is convinced it works and accepts P, plain modus tollens will force them to reject Z. This is a problem because if this is what happens, the magical strategy gives an argument: "P - therefore by my magical strategy, Q. But if Z were true, then by my magical strategy, Q would be false. Therefore, Z is false". Apply this to the case in question and it turns out that JCD can save his argument against this reductio only by denying Z. But to save an argument for life after death by assuming life before birth will not be a convincing strategy to many.

As for point 2, I wouldn't have much to add. It is perhaps metaphysically possible that BSG is true and JCD is a cylon travelling in time (and perhaps across possible worlds), but this doesn't shift the burden of the proof: he doesn't have to focus on proving that he isn't. I am still allowed to assume he isn't, unless proven otherwise. (In general, the burden shift JCD proposes would be pretty bad in court: innocent unless proven guilty would turn into guilty unless proven innocent given the possibility of guilt.)

JCD addresses ET's point 2 as follows:
As regards your second point, I do believe the burden of proof is switched. My existence is already a possibility based on the fact that I currently exist. The argument only illustrates that the burden is, was, has always been, on the other side. It's the difference between saying there will be another lion or there will be a unicorn.
Now,"I do believe the burden of proof is switched" and "The argument only illustrates that the burden is, was, has always been, on the other side." is not an adequate response, it is just a restatement of JCD's claim (actually, its strengthening). I'm not even sure if it's consistent to say that the burden is switched and that it has always been on the other side.

How about "My existence is already a possibility based on the fact that I currently exist"? Well, ET claims: mere possibility doesn't shift the burden. JCD's claim is consistent with this. I can say that existence is already based on the fact that I exist and yet maintain that this doesn't shift the burden. ET doesn't deny the possibility of JCD's existence, so stating it has nothing to do with ET's objection.

Perhaps, what JCD means is that we already know that our existence is possible (well, we do exist), whereas we don't have a compelling reason to believe that our non-existence is. If this was true, then indeed, the burden of proving the possibility of non-existence would be on JCD's opponents. Alas, the experience of the fragility of human existence is quite common, and so are prima facie consistent thoughts about death of the body as the end of life. If JCD claims human non-existence is impossible, an argument is needed.

As for point 3, JCD's reasoning goes from P to Possibly P and from Possibly P to Possibly P and Q: I exist. Therefore, it is possible that I exist. Therefore, it is possible that I exist after my body dies. Of course, the first move is valid (you don't have to conjure medievals to see that), but the second one isn't (just take Q to be not-P, for instance; or any sentence that metaphysically excludes P, if you prefer).

Tuesday, November 16, 2010

Evolutionary arguments do go astray (sometimes)

Norton (Why thought experiments do not transcend empiricism. In Hitchcock, C., editor, Contemporary Debates in the Philosophy of Science, pages 44–66, Blackwell) when talking about developing non-classical logics to model certain aspects of thought experiments gives the following, as he calls it, evolutionary argument:
I think there are some reasons to believe that no new, exotic logic is called for. In outlining the general notion of logic above, I recalled the evolutionary character of the logic literature in recent times. New inferential practices create new niches and new logics evolve to fill them. Now the activity of thought experimenting in science was identified and discussed prominently a century ago by Mach (1906) and thought experiments have been used in science actively for many centuries more. So logicians and philosophers interested in science have had ample opportunity to identify any new logic that may be introduced by thought experimentation in science. So my presumption is that any such logic has already been identified, in so far as it would be of use in the generation and justification of scientific results. I do not expect thought experiments to require logics not already in the standard repertoire. This is, of course, not a decisive argument. Perhaps the logicians have just been lazy or blind. It does suggest, however, that it will prove difficult to extract a new logic from thought experiments of relevance to their scientific outcomes – else it would already have been done! (pp. 54-55)

To gains some perspective on this argument, consider the following “evolutionary” argument. Norton, among other things, works on the philosophy of relativity. Now, relativity theory has been around, pretty much, since the same time when Mach wrote about thought experiments. So philosophers interested in science have had ample opportunity to identify and solve any philosophical issue that may be introduced by relativity theory. So, in this field, any philosophically interesting claim has already been made and any philosophically argument has been given, and Norton’s work in philosophy or relativity is redundant. Unless, of course, philosophers of science since the discovery of relativity theory have just been lazy or blind.

Tuesday, November 2, 2010

A geeky note about multilingual support for WinEdt 6.0

So WinEdt 6.0 has been out for a while now. It's a good piece of software, and I've been using various versions of WinEdt since 2003 (sometimes I use emacs with auctex, but for larger projects, I definitely prefer WinEdt).

Anyway, if you often switch between languages, you have to play around to get WinEdt to open your files with appropriate dictionaries. This is where modes kick in. It so happens that the explanation that WinEdt provides for this stuff isn't too clear, so here is a step-by-step procedure that will get you multilingual support for WinEdt 6.0.

  1. Download appropriate dictionaries. They are available here. Extract (if needed) and save the file(s) somewhere. It doesn't really matter where, just make sure you know where it is. Say you want to install Polish and download pl_huge.dic saving it at C:\Program Files\WinEdt Team\WinEdt\Dict\pl_huge.dic.
  2. Run WinEdt 6, go to Options. Under Editor: Mouse, Modes, Defaults you will find Modes. Open the file.
  3. Now, what I do, is I look at the submodes and add a line: MODE=":PL|*.pl.tex" This creates a submode PL which is assigned automatically to all files ending with .pl.tex
  4. I also replace corresponding lines for US and UK English, using: MODE=":UK|*.uk.tex" and MODE=":US|*.us.tex"
  5. That way, whenever I have a file in Polish I save it as, if I have a file in American English, I save it as, and so on.
  6. This sets the modes right. Now dictionaries. In the Options tree go to Dictionary Manager and then to Word Lists.
  7. Then you add something like this to the list:
FILE="C:\Program Files\WinEdt Team\WinEdt\Dict\pl_huge.dic"

The first line just gives your setting a name. The second locates the dictionary file (that's why you need to remember where you saved it). The mode filter tells you for which modes this dictionary is active.

8. Now you also have to turn off your English dictionaries for the Polish file and switch them on for US and UK submodes. That is, you look at all English dictionaries listed and make sure their mode filter contains appropriate submode. Thus, for example:

DICTIONARY="English (common)"

Then, you save all the changes you've made in the options sections, restart WinEdt to make sure it loads its settings anew and check if language selection works properly. Make sure the files you use have appropriate extensions.

If you think something's still off, feel free to include mode declaration in the beginning of each file. Just include % -*- TeX:US -*- (or % -*- TeX:PL -*- or...) in your preamble.