who is a scientist

T. Ryan Gregory asks this important question: Who is a scientist? It's a followup to a post titled: "Graduate students are not professional scientists. Discuss," which – briefly – argued that grad students are scientists in training, not yet scientists-full-stop. In the later post, he explains:

Here are the criteria I threw out off-handedly for the purpose of discussing the NYT story about science blogs [this one -Josh]: - Does scientific research for a living,
- Publishes research in peer-reviewed journals,
- Is funded by granting agencies to do it,
- Does not just write about it, or study it, or do some of it as a grad student, or only teach it.
This wasn’t an official or proposed definition, as indicated by the qualifier “For the purpose of this post”. Others have raised objections to one or more of these. I don’t think they are all necessary and certainly none is sufficient. So, let’s go through the exercise and think of some criteria that would distinguish a “professional scientist”. Nowhere in here is there an implication that graduate students, industry scientists, government scientists, postdocs, or anyone else doesn’t “do science” when they are engaged in research, so let’s get beyond that straw man if we can.
As I noted in the last post, lots of people want to be called “scientist”, presumably because it carries some prestige. But if anyone who does an experiment is a “scientist”, then the term isn’t meaningful at all.
So, assuming that we want the term to mean something, what makes someone a scientist?

I disagree with his means of distinguishing grad students from "professional scientists." To my mind, a professional scientist is someone who is paid to do science. Grad students are usually paid to do science (either directly, as research assistants, or through teaching assistantships awarded to support them while they conduct their research), and are therefore professional scientists. Publication in peer-reviewed literature is an important step in the process of science, but it is a lagging indicator and does not belong on the list in that form. Doing work one intends to publish, and which one presents to legitimate scientific peers for review (even through informal channels) should suffice. And I don't know why Gregory insists that the research must be funded by granting agencies. Scientists in industry or working for a government agency and supporting their research out of an operating budget are still professional scientists if they are doing legitimate science and being paid to do it. People getting funds from a granting agency but not doing legitimate science are still not scientists (e.g., homeopaths, cold fusion, astrology, woo of other sorts).
But saying that a professional scientist is someone who gets paid to be a scientist just brings us back to the closing question from that quotation: "what makes someone a scientist?"
It's helpful to consider some examples: Charles Darwin did not get paid to conduct scientific research. He had no academic degrees in science (or natural philosophy). But he was, by calling and by mindset, a scientist. He lived in an age where science was shifting from a gentleman's pursuit to a professional career option (like doctor or lawyer or parson), which makes the comparison somewhat illegitimate. Yet it also emphasizes that "professional scientist" is not an age-old category, and this debate is one that has a relatively brief and as-yet-unresolved history.
A slightly more modern example, Albert Einstein, did impressive scientific work while working as a patent clerk. The scientists who built the first atomic weapons were not grant-funded, but certainly advanced our understanding of fundamental physics as they were hurtling the world toward a fearsome future. The datasets generated by volunteer birders through Breeding Bird Surveys and Christmas Bird Counts are invaluable resources for ornithologists and ecologists. David Attenborough's films have captured new animal behaviors and brought many people to a deeper interest in science and natural history. He and his production crew have become expert in natural history and biology, and have developed new tools to investigate and record that behavior. A whole field of hobbyist biotechnology is taking off in rented lab spaces around the world. A web software developer spends his spare time tinkering with what may turn into the future of nuclear fusion. Then again, many of the professors who get grants and publish papers spend so much time administering their lab and writing grants and advising students and preparing lectures that they may not conduct any actual experiments (though they do observe and supervise the students and postdocs who are performing the research funded by those grants). Which of these folks are scientists?
An auto mechanic or a plumber might apply the scientific method to fixing your car or your toilet, but neither they nor anyone else thinks that makes them scientists. They aren't trying to increase our knowledge of the world, to derive general truths from their experimentation, and that's the goal of science.
This also eliminates some but not all of the birders who contribute to CBC and BBS bird occurrence datasets. Being able to recognize a birdcall and check a bird off your life list is not doing science. Making systematic observations about the world is a key part of science, but if the goal is not to test hypotheses and to use that to generate some sort of synthesis, it isn't science. But some birders, and some hunters and anglers, do make observations. They notice shifts in coloration or migration, and try to catalog and explain those shifts. They may lack the formal training and structure necessary to present their results at conferences or in journals, but I'm unwilling to say that they are not doing science, to say that they are not scientists in those moments. We'll call this science as an activity. It's something someone can do or not do, not a defining personality trait. It does not label the person, only the activity.
Darwin and Einstein were scientists wherever they went, whatever they did. Their mind were inexorably wired to constantly strive after organizing principles, and to find ways in the most mundane observations to test those principles. This is science as a mindset – science as avocation/vocation. I think David Attenborough shares that mindset, as does Richard Dawkins, though neither is currently pursuing a career in science. Dawkins shifted from employment as a biologist to become Professor for Public Understanding of Science in 1995, and has since retired to run a nonprofit foundation devoted to public advocacy on behalf of skepticism, atheism, science, and rationalism; all valid goals but not inherently scientific actions.
A successful grad student must be someone who pursues science in that latter sense. Some people leave graduate school because they do not pursue science in that sense (they are not scientists in that sense and do not want to become scientists in that sense), and others leave because they don't want to pursue science as a profession (they are not interested in science as a career, though they have the mindset). Senior professors, the ones spending all their time administering a lab and serving on panels and so forth, are scientists in the sense that science is their calling, even though they spend little time doing the hands-on part of the scientific work.
Some people with no scientific training still manage to gain the knowledge necessary to participate actively in the scientific enterprise. The amateur fusion researcher is taking ideas that had been kicked around by professionals in the field, and is slowly testing those ideas in his own time. A fusion scientist of my acquaintance responded to the article linked above about a web developer building and testing a new fusion reactor:

He's probably not achieving anything significant scientifically - but what a great hobby and learning experience for him, and what a great plug for fusion in general! Science, the olde fashioned way.

He's doing science. Not huge science, perhaps, but breaking new ground. To say he isn't a scientist because he isn't grant-funded or hasn't got a doctorate or is unlikely to publish in major journals seems to trivialize what science is, to make a PhD into a license to perform science rather than certificate of scientific accomplishment. And that impoverishes those with PhDs, those seeking PhDs, and those without PhDs who still want to and do contribute to the scientific effort. So who is a scientist? I'd say: someone who seeks to expand the collective limits of knowledge about the natural world by developing generalized claims and testing them against empirical evidence. Science is an aspiration as much as an accomplishment, hence "seeks." Expanding our "collective" limits means that it is necessary to communicate results to other relevant experts, and that those experts agree that new findings are interesting and correct. I don't think a new occurrence record for a bird constitutes "science" in and of itself (though discovering a new species does), because new occurrence records to not, on their own, test some generalized principle (some such observations do test a hypothesis, and that's a different story). Some scientists do that with thought experiments and observation, others through meticulous lab work, and others by watching stones sink into the sward of a country estate.