Most conspiracy theories are simply theories, nothing more. One could have changed my life, but didn't.
I was a mathematics major as an undergraduate. Most mathematicians live absurdly boring lives. They spend days split between teaching undergraduates calculus, teaching about one slightly more sophisticated class a week, and sitting in a room with a blackboard and a pad of paper writing a five page paper or two every year whom no more than one or two other people at your university can appreciate. For the most part, nobody cares about your work but a very elite group of global colleagues.
But, certain kinds of mathematicians write papers that don't show up in journals. When mathematicians gather, you hear about them. Their papers were classified for national security purposes. If this is your academic speciality, you have basically two choices, you work for the federal government, or you teach at a high school or a college where scholarship doesn't matter and you keep your papers in a desk drawer like some Stalin era Soviet novelist.
The field isn't what you might expect. The mathematics that matter for weapons of mass destruction turn out to be too elementary to regulate. Rocket science relies mostly on mathematics that were in place before President Lincoln died. The mathematics that is necessary to build a nuclear bomb was mostly in the public domain before it occurred to anyone to try to classify it. The mathematics that go into chemical weapons is both mostly elementary and to the extent that it is more sophisticated, has too many commercial applications to keep under wraps.
The agency that allegedly tries to keep knowledge under wraps in the mathematics field is the NSA, and the applications it is worried about are in cryptology. The mathematics that this relies upon, abstract algebra and number theory, for example, doesn't have a lot of practical applications outside cryptography and related fields that need to encode and decode messages, like digital telecommunications (fields where the companies involved often have government contracts anyway).
A similar issue comes up in the related field of advanced computer security. A large number of the employers in the field have defense contracts, so you need a security clearance to work for them. There may be other similar fields. While the mathematics of nuclear physics is fairly widely known, the advanced side of practical nuclear engineering is not. Biochemistry may also have some classified lines of research. Unlike computer security or nuclear engineering (which are changing so fast that ideas get stale quickly in any case), mathematicians generally aren't in it for the money, so I suspect that simple economic incentives may be less effective in approaching them than experts in some other fields.
Still, from what I heard, as I was exploring potential specializations for graduate school, most people faced with this situation sign up and shut up. Mathematicians are mostly mathematicians because they love playing with mathematical theories. Math tends not to be the most politically engaged major on campus, and joining a community of like minded people who can appreciate your work while drawing a regular paycheck with government benefits in someplace like Northern Virginia often sounds like a pretty good deal to most.
I also got the impression that most mathematicians who were prohibited from publishing and chose not the join the defense-intelligence establishment suffered quietly. Most could still find some sort of math related jobs, and some were versatile enough to find unrelated fields in which to publish. The agencies in question might keep tabs on you, but it didn't appear that they had the resources or saw the need to engage in 24/7 surveillance.
Apparently, the NSA cares more about "them" not getting knowledge that the people who would like to publish it probably wouldn't know what to do with practically in any case, than they are about keeping the flow of new discoveries going. Maybe some of the papers that pop up and get censored offer results that have already been proven "in house," so they don't need the research. Scientific and mathematical discovery is something of a sequential process. New discoveries are made when everything that needs to be known to get to that point is in place. Sometimes solitary scholars solve several key steps in isolation and seemingly come out of nowhere when their discoveries are announced. But, near simultaneous independent discoveries of the same major breakthrough, including both calculus and the theory of evolution, are more common. The NSA has had a big budget and has operated for decades in one of the most insular cultures of absolute secrecy of any of the spy agencies, so it has had time to develop the relevant fields quite a bit while suppressing domestic research until somebody abroad comes up with it anyway.
Sure, there might have been some quiet grumbling over drinks at an academic conference, or perhaps even more emphatic indignation in more private settings, but I've never heard of anyone who went public with their grievances bringing it to the popular press or writing an expose that managed to make a best seller list. Maybe somebody did and it simply never made much of an impact. Maybe mathematicians aren't good at telling their stories in compelling ways. Maybe the moral outrage that exists when political ideas are suppressed just doesn't have the same punch when you're talking about obscure issues in prime number theory that you'd have to be a mother with a PhD to love, in the face of fairly easy to understand national security concerns. If any part of this conspiracy theory falls flat, it is the absence of revelations of this one from the conspiracy theory canon that is the most problematic. One would think that someone would have talked sometime, and that someone else would have cared and gotten the story out.
Large groups of people are inherently bad at keeping interesting secrets. Mathematical theories of abstract algebra and number theory might not themselves be interesting secrets, any more than Jane Doe's income as a waitress at Moe's Diner (which the IRS faithfully guards), or the doseage of Mr. Ed's heart medicine (which the local pharmacy guards faithfully), or my client's bottom line in a tort case (which I guard faithfully as do the other employees and contractors in my small law office who could access that information). But, suppressed academic papers seem like a hot enough secret for people to care about blabbing and reporting. Math might not be sexy, but academic freedom definitely fits the bill.
Suppressed academic papers ought to be at least as interesting as alleged illegal national security wiretaps and the "echelon" program, a reputed top secret Google that searched everything and let the spooks find what they were looking for, whether the initial index search was legal or not. These two alleged NSA activities have gotten a great deal of press, presumably rooted somewhere in civic minded NSA employees who are part of the D.C. culture of whistle blowing leaks regarding questionable government intelligence activities. I'm sure that the barrage of action movies that feature the NSA or its fictional counterpart overestimate our nation's electronic intelligence capabilities (the classic implausible conceit is the big brother claim that every survellience camera and electonic device can be accessed in real time to spy on you), but that doesn't mean that the NSA's capabilities aren't formidable, if rather more boring.
Maybe those who didn't play ball were involuntarily prevented from going public, we speculated. How this would be done wasn't entirely clear. The Internet wasn't as pervasive when I was an undergraduate as it is now. Also, the work the NSA might want to suppress is distinctive enough that an anonymous posting wouldn't necessarily be very anonymous. And, the NSA is in a better position than just about anyone else to suppress something on the Internet and track it to its source, particular if the offender had personally dissed the agency's authority. They were in the cybersecurity business before the word cybersecurity existed. Covering your tracks would be a sophisticated enterprise in which you would have few people inclined to be allies for any reason more powerful than a generalized hacker's ethic. Methods of shutting up people are more ancient, but the cases of people not buying in, having any real value, and not shutting up, would have been rare enough, and mathematicians in that field are rare enough, that there might not have even been a standard operating procedure response for the few instances that might have come up. This secret, if it exists, has been well kept.
Are the rumors I heard while contemplating graduate school specialties true? I don't know. I was the rare math major whose interests were more political than cerebral. While I was something of a math prodigy in high school, it soon became clear that I wasn't the best of the best at graduate school level mathematics. I managed honorable grades in the advanced mathematics classes I took, on average, mostly in applied fields. I probably could have earned a master's degree in an applied mathematics subfield like statistics or operations research, although a PhD would have been a real challenge. But, law, with its absence of prerequisites and political overtones beckoned. A couple of years after I started hearing the rumors, I was cut off from my sources. I heard only the rarest whisper from mathematics graduate students while isolated in the separate world called law school. Maybe it was just another conspiracy theory, maybe not. But, it was assumed to be true by most of my peers when I was in college, and I've rarely heard this issue broached one way or the other since then.
Still, a pattern and practice of NSA suppression of abstract mathematics papers, at least, makes a more plausible story than much of a government conspiracy genre, and nothing raises goosebumps like plausibility.
Did I mention that one of the few aspects of the Bush Administration's war on terrorism doctrines, which the Obama administration has not repudiated, is the state secrets privilege? Hmmm. They say you have a lot of secrets revealed to you when you become President and that this can change your point of view on the matter.