So in the last few days Michael Egnor of the Discovery Institute released 8 questions for the gnu atheists that he really really wants answers to. This has already gotten some treatment from Common Sense Atheistm, PZ Myers and the always-brilliant Jesus and Mo. However, it also reminded me of another list of questions that I filed away a few months ago. So, here they are side by side:
| Michael Egnor’s 8 questions (What DO the New Atheists Actually Believe? | 10 Questions Every Intelligent Atheist Must Answer |
|
|
I filed away the 10 questions on the right to answer some of them at least as another way of exploring just how loaded and disingenuous these “stumpers” always (in my experience) are. But now that there are two sets, I just can’t do it — in fact I don’t think I can bother with such lists except to scan through them and get very annoyed at the ridiculous presuppositions.
I think nothing epitomises the faux-deep-and-meaningful question as the “why is there something rather than nothing?” spiel. I absolutely love that some people think it’s incredibly profound, sometimes even the most profound question there can be. To me it’s a clearly malformed and hence meaningless question — perhaps I’ll expand on this at a later stage.
Until then, to look at the left side, #1-#4 are I think all malformed and may not even be meaningful. #8 rests on presuming the truth of much of Christianity so the only interesting ones are #5-7. However I think #5 and #6 are very related, possibly two sides of the same coin. However, this is something nobody can explain satisfactorily so it doesn’t differentiate between atheism and any other religious position. Even #7 may be malformed: what would a world where there was a “Moral Law” look like?
Onto the right side. #1, #4, #6 and #9 are either misunderstandings or super-loaded or of the “when did you stop beating your wife?” form. #2, #5, #7 and #10 seem to contradict much of what we know about logic and science (eg. how do you self-report how logical you are without bias? how can your trust in science be based on science?). The interesting ones are #3, #8. #3 I don’t know much about — unless I facetiously answer “from the brain”. And for #8 I’m very happy to provide an answer: if all Christians agreed the Bible was allegorical it wouldn’t change the arguments for my position (that there is no reason to accept any form of theism) one bit. Believing that YHWH is an allegory would stop Christianity from being theism.
So there, I think these question series are largely designed to move the debate into semantics and as such it’s probably best not to spend too much time on them. Unless of course you have a meaningful, non-loaded question in which case comment away.
3 comments ↓
Hi Michael.
I agree with you that “Why is there something rather than nothing?” is a malformed question. It assumes that “nothing” is somehow a preferred “situation”. Why is that? What rule is supposed to be operating that prefers nothing?
Despite my reservations about the meaningfulness of the question, I will try to answer it. Here is my view (which is going to be ontologically a bit extreme for some people).
All abstractions of physical objects are also physical objects. That is to say, you can apply any interpretative algorithm you want to some object and generate information. That information is also an object. You can apply any interpretation to that… and so on. We are used to the idea of abstraction and emergent properties, and this just means not having any cut-off point. (The “Minds, Substrate, Measure and Value” articles I have been writing have been going towards this and the next one will discuss it explicitly.)
How much do you need to be able to apply interpretations to it to “find” new objects? Even if you assumed just one bit of data in the universe, you could apply an infinity of interpretative algorithms to it to find an infinity of other objects. These in turn could be subject to interpretation, and so on.
But why stop at one bit? An interpretative algorithm can be applied to an object, but whatever information it “reads” from that object depends on its own internal structure. In principle, an interpretative algorithm may not read anything at all: It may be abstracting nothing and finding objects in nothing. If we think abstractions are real, we should therefore view objects found by such algorithms as real.
Suppose now that nothing exists. Interpretative algorithms that could find objects without using any inputs could still find those objects, implying those objects exist. Therefore there can never be a state of nothingness, but even nothingness would imply an infinity of other objects that are abstractions of it.
We can summarize this as saying:
All abstractions of objects are real.
Reality is an abstraction of nothing.
Nothingness could never be, because it could always be abstracted to get something.
We don’t have to assume very much to get here. For example, we aren’t assuming modal realism. We aren’t assuming that all possible mathematical objects exist. We are simply accepting that our ideas of abstraction and emergence apply in a very general sense – that abstractions correspond to things that really exist.
To those who haven’t seen the ideas above before they may seem completely crazy at first but they actually make a lot more sense than that, see Paul’s series on “Minds, Substrate, Measure and Value” on http://www.paul-almond.com/.
My only question about this condensed version of the argument is that in the first part you speak of the proverbial “me” applying any interpretive algorithm I want to any object to get any abstraction. This seems fine, but in the 2nd half the “me” disappears and I’m not sure what step is used to justify the elimination. This might have to be something I’ll need to think about in my year off but it seems that whenever we talk about an interpretation we already speak of a physical system that has enough structure to generate semantic content.
Also if this were true (including the full multiverse theory) then wouldn’t everyone have to be immortal since there is always an interpretation by which the same person would continue to live. So perhaps our own death is an experiment that tests this, if our consciousness really is extinguished then something was wrong with the argument (even if we’re not there to say “it was wrong”!)
Hi Michael
When I talk about “you” being able to provide an interpretation, I am not suggesting that “you” have any important role to play here. In fact, what causes an object to exist, in this view, is that it is logically possible to apply an appropriate interpretation to find it, whether you are there to do any applying or not. The use of “you” was informal language. For example, I might say “If you compute 2+2 and 1+1+1+1 then you compute the same result: The “you” here is not an important part of the discussion.
Regarding immortality… er, yeah. This is a significant issue with multiverse type models of reality. It is already discussed a lot in the less-expansive multiverse concept of Everett’s “many-worlds” interpretation of quantum mechanics. Of course, in an expansive enough mutliverse, you don’t even need causal connection. (e.g. you die and then find out you got lucky: Aliens on the planet Zog happen to have written an AI with a personality and memories that are just like yours at your moment of death…) For now, I am not taking a firm position on this. I can think of arguments for and against it. I will admit to having some scepticism, but I am still trying to sort out how we should approach this. One issue here is measure: Getting killed could reduce it a lot. Should that matter? I can that someone could say, “Who cares what measure is?” Another issue is whether we can realy ignore the measure of the situations in which you are actually dead. I know why people think you can: You aren’t there to observe them. I have a thought experiment on this, which I will give now. I won’t give any answers with it, but I think it exposes a big issue in how we approach continuity.
Suppose you had multiple futures ahead of you – your reality “splits” somehow. Maybe a cosmology like the one we are discussing here is correct. Maybe Everett is correct (and I happen to think he is generally correct, actually). Maybe you are an AI making multiple copies of yourself. Now, suppose your future splits off into a very large number of different futures. In one of these you retain your normal faculties. In another, you are slightly “damaged” in a random way, so that your mind in this future is slightly different from your current mind with some randomness injected. Suppose another future involves you being more different still, and so on. We can imagine each of the “yous” in these futures being arranged in a line. At the right hand side, we have versions of you which are identical or almost identical to you. As we go along to the left we find increasingly damaged versions. Eventually, when we go enough to the left, they become so damaged that they are much less intelligent than you are now. We can keep going and eventually we encounter versions of you so damaged they are scarcely recognizable as you. Eventually they are so different to you that there is no reason even to associate them with you: They are just randomly built things with no cognitive abilities at all. The similarity with death and multiverse immortality arguments should be obvious here – except I blurred the boundary a bit and made the transition from live versions of you to dead versions gradual.
The question is: Before the split, what do you expect your IQ to be in the future? Do you take into account all these versions, or is there some cutoff when you say they aren’t similar enough to you to be considered as “your future”. If you include all of them, you have a problem: The most damaged versions of you don’t even need to be derived from you: Practically any object in the universe could be considered to be a “very damaged” copy of your brain. So do we eliminate silly futures? How? Who decides?
Leave a Comment