Evidently atheists haven't heard of Lysenkoism.
Or Relativity. We still have no observational proof of black holes or dark matter, and the idea that space can be curved is currently experimentally unprovable. Scientists have faith that they exist because the math for them to exist works some of the time.
All fair points, but I think it goes deeper than that.
There's at least three problems that I can think of:
The first is Hume's famous Problem of Induction.
Simple version is this: we've got all these things that we 'know' through evidence and observation, right? This is "empirical knowledge", or knowledge arrived at through what is known as "inductive reasoning". For example, we know that the sun rises every morning. We 'know' this because the evidence of our lives says that every day, we wake up, and the sun rises. (unless we're in Scandinavia, in which case we've developed all kinds of fancy scientific systems to explain the new, deviant observation)
The problem with this line of reasoning, however, is that it relies on the assumption that the future will be like the past. This, too, is an assumption we make, based on evidence (future things DO tend to happen like the past), which in turn relies on the same inherently unproveable assumption that the future will be like the past. i.e. the typical way by which we attempt to prove this assumption, which is often referred to as the Principle of the Uniformity of Nature, is nothing more than circular reasoning.
In most cases, this isn't really a problem, since aside from philosophy majors, nobody needs to spend their life fretting about stuff like "am I brain in a bottle?" or "am I correct in assuming that the future will be like the past?" However, it is an issue, it very well may be unresolvable, and if that is the case, it necessarily means that
any empirical system (like science)
must rest, at its core, on at least one proposition that is taken on faith alone.
i.e. unless a scientist can resolve Hume (and I would venture to suggest he can't), all science is, at its most basic level, a matter of faith.
The second is the Crisis of Reproducability.
A lot has been written about this in recent years, so I would assume euophorics - with their interest in science - are already acquaintance with it. But again, simple version: science is meant to reproducible; experimental results get published, then other teams work to see if they can replicate those results (and if they can't, that's a strong indicator that there was something wrong with the original team's research). This is one of the main rationales for having peer-reviewed journals, and for even having a "scientific community" at all. Yet analysis of scientific research has been showing that vast swathes of supposedly "sound" studies either have not, or cannot, be replicated. Usually this effects fields that, I would argue, aren't science at all (sociology for example), but even "real" science, like physics or sexology, is having a hard time with it.
The question then arises: can we accept scientific conclusions that have not been replicated? If we do not, then we have to a) put science on hold until we can identify all the "junk science" (which is going to take a lot of time, and involve a lot of heartache), and b) throw out (potentially) the majority of what "science" has told us for the past couple centuries (including stuff that could very well
be correct, despite the methodological problems of existing research). But if we do just soldier on with our science anyways, then it is on faith that we do so - faith that, despite the Crisis, this particular line of scientific reasoning, in front of us right now, happens to be sound and replicable (we hope).
The third is the fact that, while yes, IN THEORY, sound science will produce a chain evidence that can be checked and rechecked and shown to be sound, by each individual scientist, through direct empirical study, IN PRACTICE, nobody has the time, let alone inclination, to do so.
Science is complicated. The body scientific literature available for any given question within any given field is often considerable, and it's growing larger every day. This is why, unless they're just teachers, scientists cannot be "generalists" - they must pick a field, and a narrow range of questions within that field, to really study and understand. The science they do is, presumably, open to them, and they can see for themselves which hypothesis stand up to scrutiny, which don't, what the evidence is, etc etc.
But to borrow a famous phrase, every scientist "stands on the shoulders of giants". Yes, scientists tend to be professional and thorough. Maybe you cite a paper as a source for an assertion or assumption you make; then you check that paper's sources, and maybe you check THAT paper's sources, and maybe just maybe check the next down just to be safe. But nobody's going to check the source of a source of a source of a source of a source, not for one scientific fact you assert to be true, and certainly not for
every single scientific assumption you assert to be true.
And even if you WERE to go back, back and back and back, into the literature of the Before Times, and you read all the studies and experimental notes right on down to the beginning, it would be even more absurd to actually redo each of these experiments in order to check and see, empirically, for yourself, that yes in fact, Dr Soandso WAS correct when he got Japanese bullweevils to successfully mate while under the influence of this particular chemical compound, and yes, Dr Whosamawhatsi WAS correct when his team found that the compound in question acted on the protein receptors of Molecule Blahblahblah, and BY GOD YES, Dr Mrsvandertramp was utterly and completely spot-on right when she discovered that Molecule Blahblahblah was composed of three atoms of Element Xirps, five atoms of Element Quipups, and some scotch tape that had been nicked by the undergraduates from Nottingham.
ALL THIS TO SAY, ultimately, everyone - even the very best scientists (especially the very best scientists?) - must,
and do, on any given day, take countless "sound scientific assertions" to be true, not based on any direct empirical knowledge that they themselves have access to, but on faith alone. On faith that the scientists before them did the work, did the work well, and came to the right conclusions.
