Looking at journals isn't the only way to do it. You can take encyclopedias and biographical dictionaries, and check for citations and mentions there.
Funnily enough, one such attempt (http://en.wikipedia.org/wiki/Human_Accomplishment) finds that per capita scientific achievement seems to start declining around the 1920s or 1930s, which is also (with WWII and whatnot) the beginning of the modern university system.
I think per capita scientific achievement has declined, but I think it's because we did all the easy stuff already. We are starting to hit a wall where new discoveries increasingly require one or more of the following:
1) A lot of expensive tools. (accelerators, genomics cores, supercomputers, etc.)
2) A lot of labor. (The "armies of graduate students and postdocs" phenomenon is a product of this.)
3) Intelligence beyond the capacity of a single individual. (Supercomputers, large numbers of people over time, etc. are required to "grasp" something. Anything ending in "omics" is a perfect example of this.)
You can think of it as being loosely analogous to peak oil, though the underlying causes are different. With peak oil, it's because the resource itself is limited. With scientific discoveries, you might have an infinite ultimate supply but you have a finite horizon of vision. It's like if there actually was infinite oil deep in the earth, but we were strictly incapable of building drills longer than one mile.
Personally I think #3 is the limit we're really hitting. I sometimes doubt that we have the IQ power as a species to truly grasp the genome or to unify QM and gravity. We either don't have the IQ or we don't yet have the language or philosophical framework to think about those things.
We're not infinitely smart. Can your dog read Hacker News? Maybe aliens are watching us now and thinking "well, they're sort of clever but they're really just still animals... they only see two sides to every conflict, they can't think in more than three dimensions, and they can't hold more than seven or eight simultaneous ideas at once..."
I'm not saying academia is perfect. It needs to change. If we really are nearing our IQ limits, we need an environment that removes distractions and allows people to really focus to get that last bit of creativity and intelligence. Academia is too bureaucratic, noisy, interrupt driven, and stressful. All those are IQ-killers.
"Correlation doesn't imply causation, but it does waggle its eyebrows suggestively and gesture furtively while mouth 'look over there'" http://xkcd.com/552/
I might buy the End of Science argument, but this would be far from the first time that it has been made, and I'm not sure it really works. Look at computer science; how much of that was or could have been worked out in the absence of an actual computer? And as much computer development was aided by government funding, the commercial imperatives were just as compelling. (Look at IBM's pre-digital computer success.)
I agree... like an economic bubble, I don't think it's possible to ever "call the top" except in retrospect. But it does worry me. I don't think human beings like to think about it because it bruises our collective ego to think that there might be a bunch of stuff that we're too dumb to get.
But I do wonder if we've passed "peak innovation." The biggest reason is the 1960s. It really feels like virtually every piece of technology that we use outside of maybe biotech was invented in the 50s and 60s.
Looking back, I see two peaks. I see a peak in per capita innovation in the early 20th century, and then a second peak in absolute terms centered at approximately 1960.
Funnily enough, one such attempt (http://en.wikipedia.org/wiki/Human_Accomplishment) finds that per capita scientific achievement seems to start declining around the 1920s or 1930s, which is also (with WWII and whatnot) the beginning of the modern university system.