The Nature of Science
Posted: Thu May 23, 2013 10:36 pm
In response to the recent interest generated by the subject of the nature of science on Steve's show, I wanted to start a thread on that topic. In what follows I am going to discuss both the nature of science and the nature of the supernatural.
First, I want to say that there are many issues to sort through that are related to the nature of science. This is especially true for Christians, as we sometimes feel that the culture views us as being at odds with the discoveries and proclamations of the scientific community. Or, for the good many Christians who are scientists and sometimes leaders in their various fields (Francis Collins is a popular example), there is unavoidably a question of how to integrate or adjudicate the supposed problem of having two authoritative sources of knowledge --science and scripture. Amongst the various concerns that we could address, what I specifically want to focus on here is the following question: What is science? Since I take this to be at the center of so many points of controversy, I think that it would be helpful to begin trying to answer this question in a way that might lead us to resolve some of the problems that extend from a lack of clarity on the nature of science.
Let us be clear then about what we are asking for. In the circles that discuss the nature of science under the academic heading of the philosophy of science (a sub-discipline of academic philosophy), this question roughly translates as the demarcation problem, which asks: How do we distinguish genuine science from non-science? In order to answer this question, philosophers typically go about offering a definition of science. It is relevant to observe, at this point, that it is the job of philosophers to define science and not the job of scientists, though I recognize that this is a controversial statement. It will suffice to say, for my purposes, that the question "What is science?" cannot be answered by any scientific investigation and can only be addressed by scientists when speaking philosophically--that is, as philosophers. As such, philosophy is actually authoritative over science in this way (and in others) because it is the job of philosophy to investigate and provide a definition of science (if one can be found). There is much more that could, and in fact should, be said on this point, but it would distract from my main goal.
To demarcate science from non-science is to provide a set of necessary and sufficient conditions by which a thing can be considered "science" or "scientific." A necessary condition is a requirement that something must fulfill in order to be science, and a sufficient condition occurs when we have stated all of the necessary conditions (or all of the individually sufficient conditions) for something to be science. One fairly uncontroversial necessary condition for something to be science is that it involve observational evidence. A sufficient condition for something to be science is that it be published in a scientific journal (though this of course doesn't specify what counts as a "scientific" journal and in that sense begs the question, though we can ignore this for the time being).
A caller to the show (I think his name was John) was talking about the nature of science and in the course of things he ended up advocating three different definitions of "science" without quite realizing it. Here are the three definitions he offered:
1. Science is an investigative methodology.
2. Science is a set of assumptions about the world.
3. Science is an attitude.
Definition (1) is the most popular because people usually talk about "the scientific method" as if there were a clearly identifiable methodology that was unique to the sciences. This includes things like proposing hypotheses, setting up experiments, gathering data, evaluating data in the light of one's hypotheses, and then drawing conclusions. The problem with this definition of science is that it doesn't do what most scientists want it to do. One important thing it doesn't do is that it doesn't rule out supernatural explanations for phenomena that are observed during the process. There is nothing about prescribing this method which rules out what kinds of causes I am allowed to appeal to when drawing conclusions--whether natural, supernatural, or otherwise. It also ignores the history of science in which many "scientific" conclusions were arrived at by methods that leave out at least one of these steps. Lastly (to state one problem among many more that could be included), it doesn't state anything about the goals of science. Does science try to provide explanations or models of understanding? What about the discovering of natural laws and how they feature in scientific accounts of the world? If science is simply a method, it seems that one could do a lot of different things with a method we wouldn't want to call all of them "science."
The second definition usually appeals to ideas like the claim that all natural phenomena have natural explanations, or that nature is a closed system, or that the best explanations appeal to natural causes and laws. This view also has some rather obvious short-comings. First, what are we supposed to do with these assumptions? Write them down in our personal journals and read them as devotionals? The point here is that it makes no mention of science as an attempt to learn something about the world. Rather, to go this route is to turn science into a set of beliefs about the world that a person must hold in order to do science. Many people implicitly assume this definition when they say that science "looks for the best natural explanation" without realizing that they are requiring that one make assumptions about the nature of the world in order to do science, yet I doubt that they would want to deal with the consequences of assuming such a view.
The last definition was only brought up in passing, but I have heard people unintentionally advocate it by saying something like this, "What science does it is says that we need to not stop looking for the explanation of something just because we can't find one. Science keeps trying to look for a natural explanation in terms of natural laws rather than assuming that God did it or that we just can't know certain things." This is all well and good but the person who says this doesn’t realize that they are actually advocating the definition of science as an attitude, not an investigative methodology nor even as a set of assumptions about the world such as what kinds of explanations are the best. In order to be doing "science," all we have to do is have this attitude.
Now, I realize that people don't entertain these last two definitions of science as serious candidates for distinguishing science from non-science. And the reasons for why should be obvious because they are neither necessary nor sufficient conditions for doing science. That is, one can do science without believing that natural phenomena always provide the best explanations, and also without the rather chipper attitude (which I don't necessarily have a problem with) of "We'll find an explanation if we just keep looking!" And as for the first notion referring to "the scientific method," I have hopefully said enough about this view along similar lines of criticism.
Without going too much longer on this point, I will forego further discussion of demarcation (as I trust that people can bring it up if they want to) and simply state my own view. I think that there have been no successful demarcation criteria proposed to date. This means I think that there are no hard and fast rules about what count as "science" and "non-science," or "scientific" and "non-scientific." In addition, I believe that the question of defining science naturally raises the question of scientism --the view that only science provides true knowledge of the world--and actually shows that there are many sources of knowledge besides science which we can use to learn about the world. One of these sources of knowledge is scripture. Another is reason or logical intuition, upon which much philosophical reflection is based. In fact, I think that philosophy as a discipline provides conclusions which can correct those arrived at by the scientific community, because science itself requires certain philosophical assumptions and as such is incapable of justifying those assumptions by appeal to "scientific" investigation. Science is but one method of learning about the world, and it is not even the best method at that.
Now I want to briefly address the question of the supernatural, which was also raised by the caller to Steve's show. It is often said that supernatural explanations are in some manner "unscientific" because they appeal to non-natural, unobservable causes. This is said to lead to intractable problems such as the explanatory equivalence that exists between competing supernatural explanations for a given "supernatural" event (the resurrection of Jesus comes to mind). This, in turn, is supposed to cast doubt on our ability to know anything about the supernatural because, if we cannot directly observe it, then we cannot know that, say, God raised Jesus from the dead rather than Zeus or some other deity. I will respond to these each of these points briefly. First, supernatural explanations are only "unscientific" insofar as methodological naturalism is the reigning paradigm of scientific investigation. Methodological naturalism requires that scientists only cite natural causes for phenomena that they investigate when drawing conclusions about them in "scientific" publications. Independent justification of this kind of naturalism is required beyond the fact that "it's what scientists do," and therefore one cannot look to the sciences for support for the claim that supernatural causes are "unscientific." They are "unscientific" because scientists do not allow themselves to appeal to them, not for any reasons independent of assumed scientific practice. Second, supernatural explanations may in fact face difficulties of explanatory equivalence, but this same problem besets many scientific theories. Anti-realists about atomic phenomena claim that atoms, subatomic particles, and the like do not exist but are convenient fictions that help us make predictions about the world. The problem for scientists is that they cannot directly observe atomic or sub-atomic phenomena but must observe their effects, so they cannot disprove the anti-realists. We are thus faced by an explanatory equivalence in that the data is equally accounted for by realist and anti-realist views of very small physical phenomena. Therefore, explanatory equivalence is not a sufficient reason for disqualifying supernatural explanations. Lastly, the context of a supernatural event (an event in the natural world that is caused by a supernatural entity) often provides a means of gaining knowledge about the cause. Thus, if Jesus predicted his death and resurrection and was casting out many demons, then it is unlikely that Satan would have raised him from the dead (as it would have been rather counterproductive). We then have a reason to believe that it is more likely that Jesus and the supernatural being(s) that would endorse him were involved in his resurrection rather than those who would oppose him, as a resurrection seems like a vindication of one's claims.
In any event, I have written enough by now and need to stop.
First, I want to say that there are many issues to sort through that are related to the nature of science. This is especially true for Christians, as we sometimes feel that the culture views us as being at odds with the discoveries and proclamations of the scientific community. Or, for the good many Christians who are scientists and sometimes leaders in their various fields (Francis Collins is a popular example), there is unavoidably a question of how to integrate or adjudicate the supposed problem of having two authoritative sources of knowledge --science and scripture. Amongst the various concerns that we could address, what I specifically want to focus on here is the following question: What is science? Since I take this to be at the center of so many points of controversy, I think that it would be helpful to begin trying to answer this question in a way that might lead us to resolve some of the problems that extend from a lack of clarity on the nature of science.
Let us be clear then about what we are asking for. In the circles that discuss the nature of science under the academic heading of the philosophy of science (a sub-discipline of academic philosophy), this question roughly translates as the demarcation problem, which asks: How do we distinguish genuine science from non-science? In order to answer this question, philosophers typically go about offering a definition of science. It is relevant to observe, at this point, that it is the job of philosophers to define science and not the job of scientists, though I recognize that this is a controversial statement. It will suffice to say, for my purposes, that the question "What is science?" cannot be answered by any scientific investigation and can only be addressed by scientists when speaking philosophically--that is, as philosophers. As such, philosophy is actually authoritative over science in this way (and in others) because it is the job of philosophy to investigate and provide a definition of science (if one can be found). There is much more that could, and in fact should, be said on this point, but it would distract from my main goal.
To demarcate science from non-science is to provide a set of necessary and sufficient conditions by which a thing can be considered "science" or "scientific." A necessary condition is a requirement that something must fulfill in order to be science, and a sufficient condition occurs when we have stated all of the necessary conditions (or all of the individually sufficient conditions) for something to be science. One fairly uncontroversial necessary condition for something to be science is that it involve observational evidence. A sufficient condition for something to be science is that it be published in a scientific journal (though this of course doesn't specify what counts as a "scientific" journal and in that sense begs the question, though we can ignore this for the time being).
A caller to the show (I think his name was John) was talking about the nature of science and in the course of things he ended up advocating three different definitions of "science" without quite realizing it. Here are the three definitions he offered:
1. Science is an investigative methodology.
2. Science is a set of assumptions about the world.
3. Science is an attitude.
Definition (1) is the most popular because people usually talk about "the scientific method" as if there were a clearly identifiable methodology that was unique to the sciences. This includes things like proposing hypotheses, setting up experiments, gathering data, evaluating data in the light of one's hypotheses, and then drawing conclusions. The problem with this definition of science is that it doesn't do what most scientists want it to do. One important thing it doesn't do is that it doesn't rule out supernatural explanations for phenomena that are observed during the process. There is nothing about prescribing this method which rules out what kinds of causes I am allowed to appeal to when drawing conclusions--whether natural, supernatural, or otherwise. It also ignores the history of science in which many "scientific" conclusions were arrived at by methods that leave out at least one of these steps. Lastly (to state one problem among many more that could be included), it doesn't state anything about the goals of science. Does science try to provide explanations or models of understanding? What about the discovering of natural laws and how they feature in scientific accounts of the world? If science is simply a method, it seems that one could do a lot of different things with a method we wouldn't want to call all of them "science."
The second definition usually appeals to ideas like the claim that all natural phenomena have natural explanations, or that nature is a closed system, or that the best explanations appeal to natural causes and laws. This view also has some rather obvious short-comings. First, what are we supposed to do with these assumptions? Write them down in our personal journals and read them as devotionals? The point here is that it makes no mention of science as an attempt to learn something about the world. Rather, to go this route is to turn science into a set of beliefs about the world that a person must hold in order to do science. Many people implicitly assume this definition when they say that science "looks for the best natural explanation" without realizing that they are requiring that one make assumptions about the nature of the world in order to do science, yet I doubt that they would want to deal with the consequences of assuming such a view.
The last definition was only brought up in passing, but I have heard people unintentionally advocate it by saying something like this, "What science does it is says that we need to not stop looking for the explanation of something just because we can't find one. Science keeps trying to look for a natural explanation in terms of natural laws rather than assuming that God did it or that we just can't know certain things." This is all well and good but the person who says this doesn’t realize that they are actually advocating the definition of science as an attitude, not an investigative methodology nor even as a set of assumptions about the world such as what kinds of explanations are the best. In order to be doing "science," all we have to do is have this attitude.
Now, I realize that people don't entertain these last two definitions of science as serious candidates for distinguishing science from non-science. And the reasons for why should be obvious because they are neither necessary nor sufficient conditions for doing science. That is, one can do science without believing that natural phenomena always provide the best explanations, and also without the rather chipper attitude (which I don't necessarily have a problem with) of "We'll find an explanation if we just keep looking!" And as for the first notion referring to "the scientific method," I have hopefully said enough about this view along similar lines of criticism.
Without going too much longer on this point, I will forego further discussion of demarcation (as I trust that people can bring it up if they want to) and simply state my own view. I think that there have been no successful demarcation criteria proposed to date. This means I think that there are no hard and fast rules about what count as "science" and "non-science," or "scientific" and "non-scientific." In addition, I believe that the question of defining science naturally raises the question of scientism --the view that only science provides true knowledge of the world--and actually shows that there are many sources of knowledge besides science which we can use to learn about the world. One of these sources of knowledge is scripture. Another is reason or logical intuition, upon which much philosophical reflection is based. In fact, I think that philosophy as a discipline provides conclusions which can correct those arrived at by the scientific community, because science itself requires certain philosophical assumptions and as such is incapable of justifying those assumptions by appeal to "scientific" investigation. Science is but one method of learning about the world, and it is not even the best method at that.
Now I want to briefly address the question of the supernatural, which was also raised by the caller to Steve's show. It is often said that supernatural explanations are in some manner "unscientific" because they appeal to non-natural, unobservable causes. This is said to lead to intractable problems such as the explanatory equivalence that exists between competing supernatural explanations for a given "supernatural" event (the resurrection of Jesus comes to mind). This, in turn, is supposed to cast doubt on our ability to know anything about the supernatural because, if we cannot directly observe it, then we cannot know that, say, God raised Jesus from the dead rather than Zeus or some other deity. I will respond to these each of these points briefly. First, supernatural explanations are only "unscientific" insofar as methodological naturalism is the reigning paradigm of scientific investigation. Methodological naturalism requires that scientists only cite natural causes for phenomena that they investigate when drawing conclusions about them in "scientific" publications. Independent justification of this kind of naturalism is required beyond the fact that "it's what scientists do," and therefore one cannot look to the sciences for support for the claim that supernatural causes are "unscientific." They are "unscientific" because scientists do not allow themselves to appeal to them, not for any reasons independent of assumed scientific practice. Second, supernatural explanations may in fact face difficulties of explanatory equivalence, but this same problem besets many scientific theories. Anti-realists about atomic phenomena claim that atoms, subatomic particles, and the like do not exist but are convenient fictions that help us make predictions about the world. The problem for scientists is that they cannot directly observe atomic or sub-atomic phenomena but must observe their effects, so they cannot disprove the anti-realists. We are thus faced by an explanatory equivalence in that the data is equally accounted for by realist and anti-realist views of very small physical phenomena. Therefore, explanatory equivalence is not a sufficient reason for disqualifying supernatural explanations. Lastly, the context of a supernatural event (an event in the natural world that is caused by a supernatural entity) often provides a means of gaining knowledge about the cause. Thus, if Jesus predicted his death and resurrection and was casting out many demons, then it is unlikely that Satan would have raised him from the dead (as it would have been rather counterproductive). We then have a reason to believe that it is more likely that Jesus and the supernatural being(s) that would endorse him were involved in his resurrection rather than those who would oppose him, as a resurrection seems like a vindication of one's claims.
In any event, I have written enough by now and need to stop.