I think the question of why we don't take Last Thursdayism seriously is an interesting one. Our whole approach to reality is based on generalization - of viewing our local situation as being a special case of something more general. I discussed Last Thursdaysim in an article I wrote previously, part of which argued that we should think we live in a multiverse, and that refusing to accept this is little more than a weaker version of Last Thursdayism. I will show what I said about Last Thursdayism, because it should go some way to answering the very reasonable question being asked here:
A useful example of the kind of thinking in which someone restricts his world view is provided by the idea of “Last Thursdayism”. [51] This was a world view created by Seanna Watson in a Usenet post in 1992 as a parody of similar views being proposed by theists. Last Thursdayism claims that the world was created last Thursday, with everything in a state that made it seem older – for example, with layers of sedimentary rock already laid down, growths rings in trees and humans brains containing false memories of events before last Thursday.
Last Thursdayism seems absurd. It is supposed to seem absurd. We should ask, however, what is so wrong with it? Why should we think that the world we know does not just extend, temporally, as far back as last Thursday or some other recent time? (Clearly, we are not seriously proposing this, but rather we are trying to get at how we justify extending reality.) If you were arguing with a “Last Thursdayist”, what could you say?
You could say that lots of things around us in the world are evidence of a longer past – for example, that our memories of events before last Thursday are evidence of events before last Thursday, or that sedimentary layers of rock are evidence of a still more distant past. The Last Thursdayist, however, could merely respond by saying that these things are not evidence of anything unless you assume that there is some more general pattern, extending back in time beyond last Thursday, of which we are a part – that, instead, what has happened since last Thursday is a special case.
You might point out that it is implausible that things should appear with a “fake” history, but the Last Thursdayist could respond by saying that things do not even have a fake history, really: instead, they only appear to have a fake history if you insist that we are part of this bizarre “general pattern” which extends beyond last Thursday.
You might ask for an explanation of how things came into existence without any cause, but the last Thursdayist could insist that time just goes back as far as last Thursday, that that is the temporal limit of reality, and that there is no reason to think that anything exists beyond this.
You could ask why things happened to be in this specific state at the start of time: why did growth rings in trees exist, or layers of sedimentary rock exist, or complex structures in brains that would normally indicate previous memories? The Last Thursdaysist could reply that that is just how things are – that we live in a reality with a temporal edge, that the temporal edge is the moment the world started to exist last Thursday, and that the temporal edge just happens to have things at it with various properties.
The point here is that you will dismiss everything that the last Thursdayist says, because the very way in which you deal with the world requires you to generalize. You regard everything that has happened since last Thursday as a special case of something else that extends beyond last Thursday.
This does not mean that you must accept, without reason, that a specific “something else” exists, extending beyond last Thursday, without reason. Once you accept that what has happened since last Thursday is part of a more general pattern, you can then consider all the ways in which that pattern could be described – and the proportion of this set of patterns with some feature tells you what the probability is that the more general pattern has that feature.
We consider Last Thursdayism absurd because we generalize and view our immediate situation as part of something more general, but this is nothing more than what we are proposing here. The kind of thinking that would lead you to reject last Thursdayism should also lead you to accept the ontology being proposed here.
Some people would reject what we are saying here – that your local reality must always be part of a more general pattern, and that this can always be said no matter how extensive your local reality is – but in this case we would suggest that there would be a bias here. We reject Last Thursdayism, and other silly ideas that might be suggested, because we are used to accepting the extension of things in time and space. To reject Last Thursdayism, and yet to think that there are limits on reality, is to deal with reality in an arbitrary, biased way – where “common sense” and familiarity is telling you what exists and what does not exist. We suggest that this would be a flawed way to approach the world.
We have been discussing formal descriptions of the structure of relationships that extends beyond your perceptions, and such a formal description could be considered to be an algorithm (although we could change this position if algorithms were found to be inadequate). You can never consider the full reference class of all the possible structures in detail. You will always have to limit the size of the reference class by placing limits on the lengths of descriptions.
The justification of the idea of an infinite multiverse in which a kind of modal realism is the case is not dependent on any specific observational evidence of the world. That makes it effectively an “ontological argument”. [35] Ontological arguments are generally used to attempt to prove the existence of God – something which we do not regard as at all viable – and are often treated, rightly, with suspicion. In this case, however, the argument is sound. We can make an ontological argument for the existence of some state of affairs if we can show that claims to the contrary are infinitely specific. Once that is established, “evidence” is redundant and irrelevant.
and another part of the discussion:
The situation that we are in, if we are asked to provide “evidence” of the kind of multiverse that we are proposing, or if we try to find evidence that will persuade ourselves, has some similarity with the situation in which we would be if we had to provide evidence to an adherent of “Last Thursdayism”, the parody theory mentioned previously, that states that the world came into existence last Thursday, complete with “fake” evidence of a longer history. [51] If someone did subscribe to that view, and asked for evidence, how could you persuade him otherwise? You could show him a tree that had taken many years to grow, but he would, presumably, simply tell you that it was made like that last Thursday. You could point to his own memories of what has happened, but he could just say that his brain was made with a structure giving him those memories already in place last Thursday – that he had fake memories like one of the “replicants” in the film ‘Blade Runner’. [38] Any empirical evidence you could offer for events that are supposed to have happened before last Thursday would presumably be dismissed.
But suppose, after talking to this person and gaining experience of how he thinks, you did think of some evidence – some observation of the world – that you could offer. (We cannot say what it is because we have no idea.) You offer this evidence and he accepts it, agreeing now that the world did exist before last Thursday.
The problem here is that it should never have come to this, and it should be hard to imagine any rational basis for accepting the evidence that you offered that would not also have served as a basis for agreeing with you earlier. You have not used “real” evidence to persuade this person: instead you have just used a trick.
Similarly, there is not going to be any “real” evidence to persuade anyone that the view being proposed here is correct.
There is one role that “evidence” might play: it might bring someone’s attention to the arbitrariness or ad hoc nature of alternative views. However, we should be careful how we should view such “evidence”: it would not really be “evidence”.
EDIT - I will also give some background on Last Thursdayism from the same article:
Seanna Watson apparently created “Last Thursdayism” as a parody of the “Omphalos hypothesis” – the name given to a theistic argument in an 1857 book with the same name by Philip Gosse, that the world was created in a state that made it appear older. [30] The same kind of argument is used by some creationists today, when they claim that we can see distant stars, even though (according to them) the universe is so young that the light from them should not have had time to reach us, because the light was created in space “almost here”.
If I ever start making posts like "On the banning and partial banning of words!" then I view my life as less than worthless and I hope that my friends here would have a collection to pay for ninjas to be sent to my home to kill me*. (*=humanely)