Humans almost seemed evolved to arrive at truth through a communal effort. But if that is indeed the case, then Hollywood -- and perhaps the popular imagination -- seem rather oblivious to it.
In the movies, the scientist is most often portrayed as a lone practitioner of his or her discipline. The reason he or she arrives at new insights is reduced to the simplest explanation possible, "Scientists are smart".
But in the real world, it's not just one person -- no matter how smart -- that is so often instrumental in producing reliable results; it is rather a community of people working together. I am convinced that is one of the most important components of science. So important that, on my wildest days, I would almost -- almost -- argue that a single person working entirely alone cannot produce reliable science. Almost, but not quite.
If you take an alarmingly sober view of our brains, they are like fun house mirrors. That is, they tend to distort reality. Built into our brains are almost innumerable cognitive biases. In addition to those biases, there is also the epistemic problem that our senses do not mirror reality, but rather interpret it. The net effect is that even the smartest among us is subject to erroneous conclusions.
The fact of that can be seen in the history of thought before the rise of science. Even the most brilliant of the ancient thinkers made mistakes, and those mistakes were often enough never corrected. Thales, for instance, had the insight to demand a natural explanation for natural phenomenon, but also went on to pronounce that water was the essential element of all things. No one really got what the chemical elements were until the rise of the scientific methods.
If scientists reached reliable results simply because they are smart, then there would be less need for the various scientific methods. Those methods are vitally geared towards uncovering and eliminating cognitive biases and other distortions. What is the point of, say, peer review if you're always right?
Now, the hypothetico-deductive model of scientific inquiry, along with its many variations, is what is usually meant by "the scientific method". And that model -- in most variations -- provides a means whereby a scientist can reach reliable results while working alone, and not as part of a community of scientists.
So, I wouldn't say that science theoretically requires a community of scientists, but only that in practice, a community is by far the best option. Indeed, I think a community has so many advantages in reality checking the work of its members that it could almost be a logical requirement for reliable science.
But there is an important caveat here: A community alone is not sufficient to produce reliable results. All too often, one sees a community of people who all agree on one or another set of falsehoods -- and who lend each other moral support for believing those falsehoods. Which is the origin of the (soon to be) famous saying, "It takes more than a village to raise a truth". In addition to the community of scientists, there must be some method of arriving at reliable fact.
At any rate, it seems striking that our most reliable means of arriving at fact is science, and that science is so very often a community endeavor. Would it be such a powerful means of arriving at fact if it were not a community endeavor? We are both so prone to erroneous thinking, and yet so often good at catching the errors of others, that I believe science must be greatly advanced by being a community endeavor.