There is an old parable — not sure if it comes from someone famous I should be citing, or whether I read it in some obscure book years ago — about a lexicographer who was tasked with defining the word “taxi.” Thing is, she lived and worked in a country where every single taxi was yellow, and every single non-taxi car was blue. Makes for an extremely simple definition, she concluded: “Taxis are yellow cars.”
Hopefully the problem is obvious. While that definition suffices to demarcate the differences between taxis and non-taxis in that particular country, it doesn’t actually capture the essence of what makes something a taxi at all. The situation was exacerbated when loyal readers of her dictionary visited another country, in which taxis were green. “Outrageous,” they said. “Everyone knows taxis aren’t green. You people are completely wrong.”
The taxis represent Science.
(It’s usually wise not to explain your parables too explicitly; it cuts down on the possibilities of interpretation, which limits the size of your following. Jesus knew better. But as Bob Dylan said in a related context, “You’re not Him.”)
Defining the concept of “science” is a notoriously tricky business. In particular, there is long-running debate over the demarcation problem, which asks where we should draw the line between science and non-science. I won’t be providing any final answers to this question here. But I do believe that we can parcel out the difficulties into certain distinct classes, based on a simple scheme for describing how science works. Essentially, science consists of the following three-part process:
- Think of every possible way the world could be. Label each way an “hypothesis.”
- Look at how the world actually is. Call what you see “data” (or “evidence”).
- Where possible, choose the hypothesis that provides the best fit to the data.
The steps are not necessarily in chronological order; sometimes the data come first, sometimes it’s the hypotheses. This is basically what’s known as the hypothetico-deductive method, although I’m intentionally being more vague because I certainly don’t think this provides a final-answer definition of “science.”
The reason why it’s hard to provide a cut-and-dried definition of “science” is that every one of these three steps is highly problematic in its own way. Number 3 is probably the trickiest; any finite amount of data will generally underdetermine a choice of hypothesis, and we need to rely on imprecise criteria for deciding between theories. (Thomas Kuhn suggested five values that are invoked in making such choices: accuracy, simplicity, consistency, scope, and fruitfulness. A good list, but far short of an objective algorithm.) But even numbers 1 and 2 would require a great deal more thought before they rose to the level of perfect clarity. It’s not easy to describe how we actually formulate hypotheses, nor how we decide which data to collect. (Problems that are vividly narrated in Zen and the Art of Motorcycle Maintenance, among other places.)
But I think it’s a good basic outline. What you very often find, however, are folks who try to be a bit more specific and programmatic in their definition of science, and end up falling into the trap of our poor lexicographic enthusiasts: they mistake the definition for the thing being defined.
Along these lines, you will sometimes hear claims such as these:
- “Science assumes naturalism, and therefore cannot speak about the supernatural.”
- “Scientific theories must make realistically falsifiable predictions.”
- “Science must be based on experiments that are reproducible.”
In each case, you can kind of see why one might like such a claim to be true — they would make our lives simpler in various ways. But each one of these is straightforwardly false.
I’ve talked about the supernatural issue a couple of times before. Short version: if a so-called supernatural phenomenon has strictly no effect on anything we can observe about the world, then indeed it is not subject to scientific investigation. It’s also completely irrelevant, of course, so who cares? If it does have an effect, than of course science can investigate it, within the above scheme. Why not? Science does not presume the world is natural; most scientists have concluded that the world is natural because that’s the best explanation for what we observe. If you are ever confused about what “science” has to say about something, just ask yourself what actual scientists would do. If real scientists were faced with a purportedly supernatural phenomenon, they wouldn’t just shrug their shoulders because it wasn’t part of their definition of science. They would investigate it and try to come up with the best possible explanation.
The falsifiability question is a trickier one, to which I will not do justice here. It’s a charge that is frequently leveled against string theory and the multiverse, as you probably have heard. People who like to wield the falsifiability cudgel often cite Karl Popper, who purportedly solved the demarcation problem by stating that scientific theories are ones that could in principle be falsified. (Lenny Susskind calls these folks the “Popperazzi.”) It’s the kind of simple, robust, don’t-have-to-think-too-hard philosophy that even a scientist can get behind. Of course, string theory and the multiverse aren’t at all the kinds of things Popper had in mind when he criticized “unfalsifiable” ideas. His bugaboos were Marx’s theory of history, Freudian psychoanalysis, and Adlerian psychology. The problem with these theories, he (correctly) pointed out, was that they told stories that could be made to fit literally any collection of data. Not just “data we could realistically acquire,” but absolutely anything you could imagine happening in the world. That’s completely different from the examples of string theory or the multiverse, which clearly are saying something concrete about the world (the ultraviolet completion of quantum gravity, or conditions in the universe far outside our observable region), but to which we have no experimental access (or almost none). Of course, there’s also the issue that the demarcation problem is a lot trickier than naive Popperianism makes it out to be, but that’s another discussion. The right strategy, once again, is to look at what actual scientists would do or are doing. When faced with difficult problems concerning quantum gravity or the early universe, they follow precisely the outlined program: they invent hypotheses and try to see which one is the best explanation for the data. The fact that the data are relatively crude (the existence of gravity and gauge theory, the known cosmological parameters) doesn’t prevent it from being science.
Noah Smith (an economist) wrote an interesting post related to the “reproducibility” question. It’s another bugaboo, often raised by creationists who want to take jabs at evolution. As a working cosmologist, I know perfectly well that not all good science requires reproducible experiments. We haven’t made a Big Bang in the laboratory — yet. Few of the folks who emphasize reproducibility would go so far as to claim that cosmology (and much of astrophysics) doesn’t count as “science.” Instead, they say things like “Oh, but in cosmology you’re comparing data to theories that are developed here in Earth in response to laboratory experiments, so it’s a more complicated give-and-take.” Yes it is! What they should admit is that all of science involves this more complicated and subtle kind of give-and-take between theories and experiments.
Nothing in our three-step definition of science refers to “reproducibility” (any more than it refers to “naturalism” or “falsifiability”). The key feature of science is that it is empirical — progress is made by comparing multiple plausible theories to actual data — rather than rationalist/logical — deriving truths from reason alone. But when it comes to collecting those data, the only rule is “do the best you can.” Sometimes we’re lucky enough to be able to reproduce conditions exactly (Noah’s “Level Four”), but often we are not. What matters is that there are data, and that attempting to account for them is how we choose between various hypotheses that would have otherwise been plausible or at least conceivable. This might mean that some scientific questions are harder to decide than other ones, but that sounds like the least surprising conclusion in the world.
Some will object that this conception of science is too broad, and encompasses not only economics but also fields like history. To which I can only say, sure. I’ve never really thought there was an important distinction of underlying philosophy between what scientists do and what historians do; it’s all sifting through possibilities on the basis of empirical evidence.
Which is not to say that every worthwhile intellectual endeavor is a version of science in some way. Math and logic are not science, because they don’t involve steps 2 or 3. They are all about figuring out all possible ways that things could be, whether or not things actually are that way in our real world.
On the other hand, things like aesthetics and morality aren’t science either, because they require an additional ingredient — a way to pass judgment, to say that something is beautiful/ugly or right/wrong. Science doesn’t care about that stuff; it describes how the world is, rather than prescribing how it should be. You may think that there are objectively true statements one can make within these realms (“killing babies is wrong,” “Justin Bieber sucks”). But whether or not they are objectively true (they’re not, in any useful sense), they’re not scientific statements, in the way that “the universe is expanding” is a scientific statement. If they were, we could imagine worlds in which they were not true at all (“killing babies is good,” “Justin Bieber is awesome”). Those would be absolutely conceivable worlds, just not the ones in which we happened to live. And the knowledge of which world we lived in would have to come from collecting some data, just as that’s how we learned the universe is expanding.
Sometimes the fact that science is not the only kind of respectable intellectual endeavor gets packaged as the statement that there are other “ways of knowing.” This is an unhelpful framing, since it could be true or false depending on unstated assumptions held by the speaker. Yes, mathematics is a different way of gaining true knowledge than science is, so at that minimal level there are different valid ways of knowing. But they are not merely different methods of getting at the truth, they are ways of getting at different kinds of truth. What makes science (broadly construed as empirical investigation) special is that it is the unique way of learning about the contingent truths that separate our actual world from all the other worlds we might have imagined. We’re not going to get there through meditation, revelation, or a priori philosophizing. Only by doing the hard work of developing theories and comparing them to data. The payoff is worth it.
Arguing about definitions is most often a waste of energy, except for the obvious fact that we need to agree on what our definitions are. Claims like “free will exists” or “intelligent design is not science” or “science and religion are compatible” depend quite heavily on what definitions we are using, so at least we have to talk about them enough to be clear.
Moshe,
but I don’t see any meaning in arguing what is the “right” definition.
There are certainly pragmatic reasons for trying to come up with meaningful and accurate definitions (and demarcations), in at least some cases. Perhaps one could at least argue for the idea of discriminating between “better” and “worse” definitions, even if the idea of The One True Definition for All Eternity is a chimera.
If you think public eduction should include science eduction, and in particular biology — does that include “intelligent design”, just because the latter’s proponents decide to call it “scientific”? Or astrology?
Should a government health-care system (or a private insurer) pay for homeopathic “medicine” (given that doing so means, among other things, less money for other treatments)?
I’m not sure how throwing up one’s hands and declaring, in good postmodernist fashion, “It’s all conventional!” allows us to deal with questions like that.
Peter, there are of course practical, political, etc. consequences for using certain terms, but I am putting this aside for the moment and trying to understand the following, more theoretical question. When one makes claims about what science is, e.g that it proceeds by falsification, or inference to the best explanation, or whatever, what is the status of those assertions? Are those empirical truths, are they definitional tautologies, or are they normative statements? I think it would make it easier to evaluate such claims if I knew what type of claims they are meant to be. I started taking for granted that those are empirical statements (in which case it seemed to me the demarcation problem is “arguing about definitions”), but I can see the story is more complicated.
Boundaries are always interesting. Speculations aren’t science, yet could science exist without them? Speculations have to precede hypotheses, yet are not them until they become them.
I know our kids will leave it all on the field.
A bold effort, appreciated – one little quibble: the relentlessly hammered notion that mathematics is not science I think is at best a hastily thought statement that needs unpacking, and at worse an unfounded cliché.
I mean, there is this little thing in maths called reductio ad absurdum..
Formulate a hypothesis within a logical system, derive conclusions, show that it leads to some contradiction with the known theorems of the logical system, therefore reject hypothesis.
Replacing “logical system” by “known empirical facts”, and “derive conclusions” by “acquire new data by making new experiments”, don’t we recover the three steps that you propose define science ?
In this view, the practice of doing mathematics is nothing else than the practice of doing science, but applied to logical systems rather than to reality. The two endeavours would then be much closer than usually acknowledged as for their methods, even though the actual object of study is not the same (reality vs logical systems).
Pingback: Cos’è la scienza? | oh tempo le tue piramidi
Thomas-
Well sure, if you exchange every important part of what you do in mathematics with what is done in science, it’s going to sound like science. :-l
Sean, this is great! A couple of years ago, I learned that all school biology texts must begin with a short section called “What is Science?” (I assume that this mandate comes from creationists eager to wall off science in some way, but never mind that. Point is, it’s there in every textbook.) The examples I’ve seen nearly reduced me to tears, they’re so WRONG. They stress the hypothesis-experiment component above all else—and this in a science where vast amounts of data collection and analysis long dominated the field—and were typically written in a way to make science sound fairly, well, repulsive.
I hope you’ll find a way to communicate your thoughts on this subject to biologists and the textbook companies who are strangling students’ interest in the field.
Those three steps remind me a lot of evolution as a process.
1. Create many organisms.
2. Place organisms in world.
3. Filter out organisms that don’t fit.
Add in some Chomsky on how big data statistics can produce models of the world with great accuracy but little human understanding: http://languagelog.ldc.upenn.edu/myl/PinkerChomskyMIT.html
And my questions are these, how important are humans to the process of science, and vice versa, how important is science to humanity (ie: outside of the resulting appliance of science through engineering), and do the answers to these questions influence how we should define science?
First, allow me to point you to a somewhat less naive version of “falsification” than the one used in this thread. In a 1953 lecture, reproduced in the collection of articles published as Conjectures and Refutations, Popper deals with the problem of how to distinguish scientific from pseudo-scientific and/or metaphysical theories—mindful of the devastating problems that the inductive approach of empiricism entails. The core of that lecture is a seven-point list of conclusions that gives a good non-technical account of what falsifiability does and does not mean.
Secondly, what Popper talks about is not the question whether a certain theory can be falsified in practice at a certain point in time; rather, he is concerned with the logical properties of theories, hence the title of his central epistemological work, The Logic of Scientific Discovery. As he says there, the methodology of science means that we may progress from problems to more interesting, general, and complex problems, in more or less this way:
1. Identify an epistemological problem (i.e. a situation in which a phenomenon of the world seems to contradict our current knowledge)
2. Put up a tentative, internally consistent explanatory theory that has greater content than previous theories, i.e. that explains all that which those theories explained but also explains some things they did not.
3. Deduce conclusions from this new theory, i.e. statements that would have to be true/false if the theory were in fact true.
4. Compare the conclusions to facts, especially those conclusions which are not derivable from current theories, and most especially those which current theories contradict.
5. If the new theory survives these critical tests which current theories (our previous best knowledge) predicted it would fail, the theory can be said to be corroborated (for the time being). If it fails the tests, it may be amended, or more generally improved upon, and be resubmitted to testing, provided the amendments have increased its content. Purely ad hoc changes in theories, designed to help explain only the observed phenomenon that led to the theory being put forward, shall not be permitted.
In this way, one can make use of the logical property of deduction of transmitting, as it were, the falsity of some of its conclusions back to (at least one of) its premises. This means that experiment, or more generally: providing evidence of this or that fact, cannot command our assent to one or another conclusion; what it can do, though, is force us to acknowledge a contradiction and deal with it. In the words of logician Mark Notturno: “[Logic] cannot force us to accept the truth of any belief. But it can force us, if we want to avoid contradicting ourselves, … to choose between the truth of some beliefs and the falsity of others”. Which may not seem much, but, as he goes on to say, “so long as we regard contradictions as unacceptable, it is really quite a lot”.
Pingback: ‘Definining Science-Where To Draw The Line Between Science and Non-Science?’ By Sean Carroll | Thinkers' Forum USA
“Science does not presume the world is natural; most scientists have concluded that the world is natural because that’s the best explanation for what we observe.”
Right. So naturalism is what you get if you stick with science and empiricism in deciding what’s real. And being empirical – testing one’s claims against intersubjective evidence – is the rational choice if you want to reliably distinguish appearance from reality. Pretty straightforward.
“If real scientists were faced with a purportedly supernatural phenomenon, they wouldn’t just shrug their shoulders because it wasn’t part of their definition of science. They would investigate it and try to come up with the best possible explanation.”
Regrettably, some major US science organizations don’t take this view but rather carve out purportedly supernatural phenomena as beyond the purview of science. The National Academy of Sciences writes “Science can say nothing about the supernatural. Whether God exists or not is a question about which science is neutral.”
Too bad, and not a little ironic, that those supposedly in the business of promoting science end up suggesting that the existence of some phenomena can only be ascertained via non-empirical means, the result being they give comfort to both faith and supernaturalism. Naturalism might really move forward were they to embrace your position but of course that’s politically impossible since they can’t afford to offend religious sensibilities.
The best types of psychotherapy do not force highfalutin theories on clients; as in good science, they also encourage clients to hypothesize their own theories for why they struggle and adapt these hypotheses when their theories do not help them in adapting to need. Donald Winnicott called the therapeutic process an experiment in adapting to need. Granted, not all therapists use these techniques.
“1. Think of every possible way the world could be. ”
So, if you think about it for just a minute, you’ll see that if this were right, no-one would ever count as a scientist. More realistically, of the nearly infinite list of possible ways the world could be, scientists seem to rely on prior theory to reduce (drastically) the set of possible ways that they will consider. This is not problematic in and of itself, but it should remind us that science is not some algorithm that sifts through the vast realms of possibility and plucks out true hypotheses. You might even say that its power derives not from what it considers, but from what it refuses to consider.
1. Think of every possible way the world could be. Label each way an “hypothesis.”
If evidence does not support a hypothesis; reject the hypothesis. If the evidence supports a hypothesis; retain it. Use proven theories to avoid wasting time by going back over what is already known to be true.
Pingback: No Joking Matter | Not Even Wrong
I actually agree with what Thomas says about mathematics/logic being somewhat closer to science than many people think, but I’m not sure that mathematics is exactly like science. I would definitely subscribe to a strong form of mathematical realism, and might even entertain Greene’s Ultimate Ensemble view, as mentioned by Amelia. It’s pretty clear that mathematical/logical principles that we have discovered are extremely useful in describing the world (think all of modern theoretical/particle physics, the Schrodinger equation and its existence in a “Hilbert Space,” and the increasing importance of mathematics in biology and other domains). Some would claim that its a good representation of the world that we have come to develop, but I find that description severely lacking. If it were some sort of fictional endeavour that we embarked on, one would have to answer why we are so highly constrained in developing it. Beyond that, the success of mathematics would seem miraculous if it weren’t in fact deeply ingrained in the fabric of reality.
I think that some scientists are uncomfortable accepting a Platonistic conception of mathematics because it might harken back to religion and mysticism. This need not be the case; if one takes seriously the importance of mathematics to the sciences, as well as its ability to describe the world in such great detail, then one can take a totally naturalistic approach to accepting the reality of mathematics. No beliefs in God or the supernatural are required. I know this idea that abstract patterns and symmetries actually exist might seem foreign to someone with a physicalist/materialist bent, but the fact that modern physics boils down to mathematical relations at the very bottom is very hard to ignore. I have always maintained that physicalism is the best viewpoint, but maybe this should be redefined somewhat to allow for the existence of mathematical/logical principles at the very foundation of reality. No gods, no demons, and no crackpot numerology are required. In my opinion, one can (and probably should, according to some criteria of indispensability/no miracles argument) take the idea of mathematical realism very seriously from a purely naturalistic stance.
Any thoughts on this viewpoint from others? I’m almost positive its fairly common in mathematics/physicist circles (Einstein, Dirac, Penrose, Bertrand Russell, and countless others) from what I’ve read, but more input is always welcome.
**By the way, there was a great discussion on this idea at the Rationally Speaking blog if anyone’s interested (http://rationallyspeaking.blogspot.com/2012/09/on-mathematical-platonism.html).
It seems to me the demarcation problem is being exploited here to justify the expenditure of effort, time and money on fruitless lines of research. The history of science has examples of correct conjectures placed in cold storage because of no empirical way to proceed,
String theory does have an unexpected massless spin-2 particle so one could but claiming that gauge theory is data that is explained by string theory is pathetic.
An introductory course, such as that by Caltech’s own Yaser Abu Mostafa (see http://work.caltech.edu/telecourse.html ) would quickly inform one that the above is a failed strategy for Machine Learning. It is amusing that this is being presented as the formula for science.
“Science assumes naturalism, and therefore cannot speak about the supernatural.”
I think perhaps, that it is worth pointing out that science acts as the naturalizing agent. There are no things which we have explained through the application of scientific methods that we still label supernatural. Electricity? Old hat. Tsunamis? Now they’re natural disasters. Old lady put a hex on you? Nope, you have germs that make you sick. “Supernatural” once referred to those things that happened in the heavens, as opposed to on Earth, but now there are no more heavens so supernatural refers to the cracks between secure explanation.
Regarding math as a science, I think it is important to mention the cardinality of the two approaches. Math starts with a core set of axioms and seeks to tease those axioms into more and more complex properties and phenomena. Most mathematicians will be working with a set of axioms that have derivative properties that are well designed to help explain the natural world, but that is not a core requisite of math. Science has the exact opposite cardinality, to start with phenomena and derive axioms/theories. They are related and even have a strong history of co-evolution (better math helps your science, better science helps your math), but they are two different frameworks that can be employed to obtain/extract/procure knowledge.
“Science is what scientists engage in” is a pretty good working definition, but only if we agree on who is a scientist and who isn’t.
The above sentence is wrong on so many levels I can’t believe it was even written. Who gets to decide? What if the ballot box is stuffed? What if a bunch of alchemists take over? Science can never ever be about a poll or group think. All too often the lone wolf who went against accepted doctrine proved to be right. Or what if the SM crowd gain a majority and vote that ST is no longer science?
The sentence also presuposses that if a scientist all of a sudden decided to work on creationism then that is therefore science. Now creationism would be quickly rejected but would a scientist working on a theory that you can create a universe by creating a black hole be science? If so then if that scientist tries to study if an advanced race using this technique started the Big Bang would that be science? It fits all of your definitions so why wouldn’t it be?
To me the article is trying to allow ST and multiverse research be classified as science but in doing so the definition of science has to be so loose that it allows almost every other crazy hypothesis.
It is difficult to enter comments from a tablet, so now I’m at a computer, I’ll expand on my machine learning comments.
The problem is that that “data” is typically noisy. Machine learning hypotheses that fit the noise (“overfit”) typically fail in their predictions. Roughly speaking what you want is the class of hypotheses with the smallest Vapnik–Chervonenkis dimension that you can get away with that could plausibly express the hidden pattern in the noisy data.
—-
On a different note, “what is data” and “what is to be explained” is itself theory-dependent. As a simple example, Kepler tried to find the patterns that governed the shape of planetary motions and the periods of the orbit, but also the diameters of the orbits. It then turned out what are known as Kepler’s three laws could be explained by Newton’s laws; but the diameters of planetary orbits were found not to be “fundamental” – at least, Newton’s laws do not explain the ratios of Earth’s to Mars’ distance from the Sun, even though Kepler thought it was very interesting and something to be explained.
(I don’t think what a set of natural laws can explain and what they cannot can be solved by philosophy and specifically by resolving the demarcation problem. )
The rule about choosing the hypothesis that best fits the data would say that “Newton’s laws + Titius-Bode law” is superior to just “Newton’s laws”, because it fits more data, but does anyone here really think so?
—
There is also to be considered that we seem to prefer hypotheses that provide causal mechanisms rather than purely descriptive laws. MOND provides “the best fit to the data” of galactic rotation curves, but do we really like it?
One can construct any definition of “science” one likes to include many types of activities. You have come up with a definition that includes a host of different endeavors, as many posters have noted. Based on your three precepts, the act of labeling activities “science” or “not-science” is not very useful.
However, there is clearly a difference between inquiries that are falsifiable and repeatable and those that aren’t. Maybe these represent different levels of science, what use be called “hard science” and “soft science”. The labels don’t matter.
Non-falsifiable, non-experimentally repeatable enterprises like macro-economics, string / multiverse theory, and climate science modelling tend to generate a lot of politics and controversy. The controversy is a result of the lower level of understanding and proof possible with these types of inquiries. Whether we choose to call them “science” or not, we should admit to ourselves that they are fraught with the probability of human corruption. The evidence of this is manifest.