(This is long, but the td;lr is that there actually is a surprising and reasonable criteria for what is a science, and it probably doesn't draw the line where you want.)
So, part of what i'm saying is that it'd be nice if people would stop dismissing other fields as "not science" :P
Are you, then, going to argue that we should call Scientology a science? If not, why not? And aren't you calling other fields "not really science" once you have done so?
I have an unambiguous definition of "science" that seems kind of obvious. What journal would people in your field aspire to publish their best research in? Of the people who publish there, what journal would they like to publish their best work in? And so on. If this chain does not lead in short order to the journal Nature, you are not a science.
By this criteria, mathematics is a scientific field. (That's the biggest downside of this classification - math is not a science. But here gets lumped in as one.) So is physics. So is chemistry. Biology. Climate science. And many more. But linguistics and sociology are not. Nor is any "social science". Nor, of course, is Scientology.
Why would that work? The reason goes back to an observation of Thomas Kuhn's. In a mature science, there is a shared paradigm, creating shared beliefs about what work is likely to be important, what work is not. Because of that paradigm, work of very different kinds can be compared reasonably fairly. (Not perfectly, but fairly enough that you can reasonably identify the top papers, and have a top journal.) Furthermore the relative unanimity from that field over time leads to people outside that field believing that people in that field actually have made progress. Eventually that can lead to a paradigm shared across that field and others, which allows a fair amount of agreement about the relative value of research in very, very different areas.
The integration of a field into the cross-disciplinary value systems shared across sciences will lead to that field's top papers becoming acceptable to broader cross-disciplinary journals. And, once that has happened, you have been fitted into a chain of relative prestige that goes right up to what is generally recognized as the top journal of all time, Nature.
This structure used to be less obvious. But in recent decades the emergence of the scientific citation index, and the measurement of journal impact factors have put precise numbers on "how prestigious is this journal" and "how broadly is this journal read". This has lead to a feedback loop making researchers more inclined to put papers into top journals, which leads to better sorting by value. And this same work has made the differing citation structures between "real sciences" and less mature fields of study abundantly clear. (A clarity that was promptly used by publishers. Now that everyone knows the relative value of given journals, prices could be raised for top ones. Giving rise to the serials crisis.)
One important disclaimer. Nothing in what I say should be taken by people in "soft sciences" as a blanket indictment of the quality of their work. Their lack of shared paradigms reflects the complexity of their subjects, and the resulting difficulty for someone taking approach A to present convincing arguments to people taking another approach B that this is a better way to look at things than approach B is. This does not mean that there is not really good work going on with both groups. But it does not result in agreement, simple externally visible signs of progress, or shared value judgements about the relative importance of different kinds of research. And outsiders don't have easy access to any sign posts suggesting what work is likely to be high quality, and what work is mostly BS.
And a final note. In my experience, what I just said is obvious to people in the hard sciences, and unbelievable to people in the social sciences. For people in the hard sciences, having a reasonably shared value system about very different research across a variety of fields is just how things are. For people in the social sciences, it doesn't make sense that two people studying radically different things will have any kind of shared value system. As a fun example, go to one psychologist, and ask for a list of the 10 most important discoveries in psychology in the last 40 years. Take that list to another random psychologist, see how few of them the second has even heard of. Repeat the same experiment with two physicists, and see how many of the first one's top 10 would be in the top 10 list for the other. (The exact order will vary, but there will be substantial agreement.)
So, part of what i'm saying is that it'd be nice if people would stop dismissing other fields as "not science" :P
Are you, then, going to argue that we should call Scientology a science? If not, why not? And aren't you calling other fields "not really science" once you have done so?
I have an unambiguous definition of "science" that seems kind of obvious. What journal would people in your field aspire to publish their best research in? Of the people who publish there, what journal would they like to publish their best work in? And so on. If this chain does not lead in short order to the journal Nature, you are not a science.
By this criteria, mathematics is a scientific field. (That's the biggest downside of this classification - math is not a science. But here gets lumped in as one.) So is physics. So is chemistry. Biology. Climate science. And many more. But linguistics and sociology are not. Nor is any "social science". Nor, of course, is Scientology.
Why would that work? The reason goes back to an observation of Thomas Kuhn's. In a mature science, there is a shared paradigm, creating shared beliefs about what work is likely to be important, what work is not. Because of that paradigm, work of very different kinds can be compared reasonably fairly. (Not perfectly, but fairly enough that you can reasonably identify the top papers, and have a top journal.) Furthermore the relative unanimity from that field over time leads to people outside that field believing that people in that field actually have made progress. Eventually that can lead to a paradigm shared across that field and others, which allows a fair amount of agreement about the relative value of research in very, very different areas.
The integration of a field into the cross-disciplinary value systems shared across sciences will lead to that field's top papers becoming acceptable to broader cross-disciplinary journals. And, once that has happened, you have been fitted into a chain of relative prestige that goes right up to what is generally recognized as the top journal of all time, Nature.
This structure used to be less obvious. But in recent decades the emergence of the scientific citation index, and the measurement of journal impact factors have put precise numbers on "how prestigious is this journal" and "how broadly is this journal read". This has lead to a feedback loop making researchers more inclined to put papers into top journals, which leads to better sorting by value. And this same work has made the differing citation structures between "real sciences" and less mature fields of study abundantly clear. (A clarity that was promptly used by publishers. Now that everyone knows the relative value of given journals, prices could be raised for top ones. Giving rise to the serials crisis.)
One important disclaimer. Nothing in what I say should be taken by people in "soft sciences" as a blanket indictment of the quality of their work. Their lack of shared paradigms reflects the complexity of their subjects, and the resulting difficulty for someone taking approach A to present convincing arguments to people taking another approach B that this is a better way to look at things than approach B is. This does not mean that there is not really good work going on with both groups. But it does not result in agreement, simple externally visible signs of progress, or shared value judgements about the relative importance of different kinds of research. And outsiders don't have easy access to any sign posts suggesting what work is likely to be high quality, and what work is mostly BS.
And a final note. In my experience, what I just said is obvious to people in the hard sciences, and unbelievable to people in the social sciences. For people in the hard sciences, having a reasonably shared value system about very different research across a variety of fields is just how things are. For people in the social sciences, it doesn't make sense that two people studying radically different things will have any kind of shared value system. As a fun example, go to one psychologist, and ask for a list of the 10 most important discoveries in psychology in the last 40 years. Take that list to another random psychologist, see how few of them the second has even heard of. Repeat the same experiment with two physicists, and see how many of the first one's top 10 would be in the top 10 list for the other. (The exact order will vary, but there will be substantial agreement.)