Friday, January 27, 2012

What Is The Scientific Method?

An Physics World blog entry linked to an audio discussion of what is meant by the "Scientific Method". You can click on the BBC link to hear the whole discussion on what it is, what it isn't, or the many variations to it.

As physicists, and scientists in general, I don't see many of us sit down and discuss this. I think we just do it and it comes as second nature because it is what we have been doing all along. It is also difficult to define because there is no one single way of doing things. In the end, Mother Nature gets to decide what's what.

But because of that, I think it is rather amusing that most of the discussion on what a scientific method is being done primarily by non-scientists. Or to put it bluntly, by philosophers. I suppose that is part of what they do. But I can't help thinking of the Feynman's quote:

Philosophy of science is about as useful to scientists as ornithology is to birds.

One could say that a discussion of the scientific method is about as useful to scientists as ornithology is to birds. But regardless of that, I find it a bit weary that people who are discussing what it is are mainly non-scientists, which are people who have not gone through, or practice such method that they are trying to analyze. We all know that there is a clear difference between studying about something versus actually doing it. You could read and study about riding a bicycle till you're old, but that doesn't mean that you can gain the skill or have a feel on riding a bicycle. One actually has to get on a bicycle, practice many, many times, fall a few times, before one gains the ability to ride one. Reading or studying about something is different than actually doing it.

So how do people who have never done scientific research have the ability to discuss what the scientific method is or is not?

Zz.

24 comments:

Scott said...

In many cases, philosophers of science (Kuhn, for example) come to philosophy of science after being real-life scientists. I am not a philosopher of science, but I've taken a bunch of grad courses with them, and invariably many of the philosophy of science PhD students are, in fact, ex-scientists.

That's not really a big deal, though. What matters, in this case, is that asking philosophers of science about the scientific method is more like asking how flight works on a bird than what it's like to *be* the bird. Sure, birds might not care about a thorough description of the aerodynamics of wings, but that doesn't mean it isn't useful (*even though* it might not be useful to birds themselves).

Einstein found much use in philosophy of science, particularly Mach, to come up with the ideas leading to relativity. Afterwards, it became not just a theoretical but a philosophical question choosing between Lorentzian theories and relativistic ones, given that the equations matched up precisely.

These days, the most interesting stuff in philosophy of science for me seems to be coming from the philosophy of modeling; now that modeling is becoming a new tool in the "scientific method," it's important to distinguish what it can or cannot actually claim. I see a lot of modeling scientists and philosophers of modeling working together on this front.

It's a mistake to think that philosophers of science cannot know science, in the same way it is to say that scientists cannot know philosophy. Both have used each to great effect, and perhaps if there were more scientists sitting down and discussing the scientific method, there would be fewer philosophers doing so.

Scott said...

p.s. I realize that message came off a bit stand-offish, and I apologize. I'd like to add that I love your blog, read it regularly, and recommend it to friends =]

cephalopoda said...

Agree with Scott wholeheartedly...Not only that but would add the likelihood that by studying, and participating in, philosophical principles, scientists, and science as applied, would benefit...

ZapperZ said...

But see, I get statements like this quite often, that philosophy can be "beneficial" to scientists. However, no one tells in how and in what way? It has become one of those things that people just give lip service to without offering clear evidence that it is actually quite useful.

Still, here are a few facts:

1. Philosophy is NOT a required subject area that most physicists have to study in becoming a physicist.

2. Most physicists are NOT consciously aware of philosophical principles, much less, using them.

3. Philosophy has no bearing in the workings of physicists practically all of the time.

Based on such facts, how would one argue that it is beneficial? I would even put it to you that in many cases throughout history, it is physics that has cause a major revolution in thinking about our universe, with Philosophy coming in AFTER the fact.

It should also be noted, per Brian Greene's statements in his TV show "Elegant Universe", that if String Theory cannot produce testable results, it is not physics, but philosophy. This is not meant to be a compliment towards String Theory!

Zz.

cephalopoda said...

Well, let's see...Why would a good, understanding and application of the philosophy of science be "useful" to scientists?

I daresay that if you cannot answer this adequately - though not in a rigorously quantifiable, mathematically provable, "scientific" way - you would benefit greatly...

What's the difference between the way problems were solved, I don't know, in the Stone Age, than today? Actually, thinking about it, perhaps the same cognitive reasoning process, in rough form, was used then, but it took many centuries before these common principles were elucidated and brought together in the scientific method...

I would say that scientists need to be more in mind of the application of research, how their results may be shared, used and applied...and who is 'allowed' or permitted to use them...What are they ultimately working towards? What's the broader purpose, and implications of their work? What's the ethics related to the scientific method?

I'd say that pursuing such lines of inquiry could be of great use to scientists as they pursue their work...

And for that matter, of what use is string theory today? None, certainly for what, 99.999% of the individuals alive on the planet?
Of what use is the theory of relativity? Certainly, it's been applied to great effect in applications that have changed the lives of perhaps millions, but is it really "useful" to the common person? Would we have gotten along just fine without it?
Why should non-scientists take an interest in such arcana? Why should the public fund and support it? How much of all this scientific research is of any use at all?

The answers, at least in part, can be found in the philosophy of science...

Or are scientists just motivated and controlled by some overwhelming biologically instinctual urge to do what they do?

Scott said...

A) In the extreme case that philosophy of science has absolutely no bearing on science, so what? Ornithologists are not pointless, even if they have no bearing on birds.

B) The philosophy of space and time has traditionally been woven very tightly with the physics. Newton, Leibniz, Poincare, Mach, Einstein etc. all drew heavily from both, and it was unclear when the philosophy ended and the physics began. Look at Einstein's use of Mach's (philosophical) principle.

C) These days, as I mentioned, philosophers and modelers are working together on the limits of the new method, and many conferences include both of them together; there's often cross-citation.

D) When scientists are interested in talking abstractly about the limits and features of the scientific method, what they are doing are philosophy of science, or some related meta-science. They don't need to be philosophers of science to do so, but ones who are *very* interested in that subject get into philosophy of science, and the place for those discussions are often in philosophy of science journals rather than traditional scientific venues.

E) Philosophy of science and "science proper," although they exist in different disciplines, are not so unrelated. When you say that mathematical, untestable physical hypothesis are non-scientific, that is a philosophical position about science. I tend to agree with it. Implicit ideas about the nature of evidence, what is or is not "valid science," whether or not a certain type of test is valid for confirming a certain hypothesis, etc., are all philosophical positions that scientists use with or without their knowledge. Whether or not making those positions clear is useful to scientists themselves is something you may disagree with, but as long as people find those questions interesting and worthwhile, they have at least the same validity as poetry and art.

You may disagree that philosophy has ever contributed to science (remember, for several hundred years, they were one and the same beast), but it doesn't invalidate its existence, and considering the two as wholly separate does a disservice to both. Science has the advantage of standing on firm philosophical grounds, and whether or not the scientist is aware of it, every day they make judgments and decisions based on their implicit notions of reductionism, the nature of evidence, and so forth. Many are very well aware of these philosophical grounds, but their main goals are not to explore those grounds; that is the purpose of a philosopher of science. Sometimes, one person wears both hats.

ZapperZ said...

It's difficult to have any kind of rational discussion when I provided you with a set of FACTS, while you gave, in return, a set of conjectures.

Note also that many of what you are describing are NOT the exclusive domain of philosophy. I can say that going to a public speaking seminar would suffice in getting scientists to convey more clearly their work to the public. I myself learn the skill at presenting physics to the public not because I'm away of any philosophy, but rather based on repeated contacts with the public that I can honed my skill at communicating!

There is a difference in doing physics, and discussing ABOUT physics. I will flat out state that in doing physics, one needs no knowledge of philosophy. Period. I haven't seen any evidence to the contrary. What you have pointed out is the practice of doing physics. There's a difference.

Zz.

ZapperZ said...

"A) In the extreme case that philosophy of science has absolutely no bearing on science, so what? Ornithologists are not pointless, even if they have no bearing on birds."

I didn't say they are pointless!

I wanted to ESTABLISH an understanding that philosophy has no bearing on how physics is practiced! One needs no understanding of formal education in philosophy to be a physicist.

Before any kind of discussion can take place, one needs to established an agreed-upon sent of FACTS. I tried to establish such facts earlier.

Zz.

Martin said...

To me there's a straightforward answer to this one, in that philosophers create the intellectual framework within which scientists (and, equally, those in the humanities) operate. But to answer your three points:

1. Philosophy is NOT a required subject area that most physicists have to study in becoming a physicist.

No. But every scientist who has ever lived has been a product of the intellectual environment in which he or she was brought up. The medieval scientist's mind was trained in accordance with the tenets of scholastic philosophy. So Aristotelian reasoning underpinned medieval science. The contemporary physicist's methods of reasoning are reliant on the work of Frege and Popper, whether or not s/he is aware of the fact.


2. Most physicists are NOT consciously aware of philosophical principles, much less, using them.

See above. Reasoning is a subconscious process but informed by education; and education is dependent on the era in which the individual was born. For instance, scientists wouldn't be starting with a null hypothesis and trying to disprove it unless Karl Popper had written about it in 1934.


3. Philosophy has no bearing in the workings of physicists practically all of the time.

I completely disagree. Concepts which to a modern physicist are second nature ('evidence', 'hypothesis', 'data', 'assumption', 'observation') are all ideas which could be defined in many different ways. The definitions *you* attach to such notions were arrived at by centuries of philosophy. Your forebears would have different definitions for these ideas. And most importantly, the whole method of scientific enquiry - do we start with a conjecture, and then look for evidence, or do we start with observation and try to deduce a cause? - the process of reasoning, is only what it is today because of philosophy.

In conclusion: philosophy is of enormous significance not just for the history of science but for contemporary science. The fact that scientists may not make conscious reference to philosophy when they experiment or publish does not diminish that fact.

ZapperZ said...

If you simply attach the relevence to something simply via "induction", then one can also argue that every scientist is also a religious person and practice religion! Why? Because if you trace the history of many scientific endeavor, it came out of such an environment. Would you consider yourself as a Muslim? You should, because after all, you used ALGEBRA and a whole slew of other Intellectual products from Muslims scholars. Using your logic, we are all practicing the Islamic religion.

If formal philosophy goes away today, it may affect how science deals with the public and may affect how the esoteric metaphysical view of our universe. But for those of us who do the "Shut Up And Calculate" view of Feynman, it won't effect our daily operations. I have seen no such evidence to the contrary.

Zz.

Scott Weingart said...

As per Martin's comment, it may not affect you in your daily operations, but it will affect the health of the future of science. We are where we are today, scientifically, not just because science has "progressed," but because its philosophy has as well: what constitutes good and bad evidence, etc. This has been a collective effort of both scientists and philosophers.

Martin said...

That's an utterly incoherent argument. I was not suggesting that the cultural context of an argument has any relevance; I'm arguing that the science of any era is defined by the mode of reasoning that underpins it. Yes, Arabic science originated new ways of thinking, but those ways of thinking have nothing to do with religion.

Philosophers originate methods of thinking. You use those methods of thinking. Ergo, you employ methods originated by philosophers. That you were taught those methods by scientists, not philosophers, does not diminish the contribution of those philosophers.

ZapperZ said...

Er.. hang on. I didn't realize we're getting into the historical aspect of such a thing. If you look my blog post, it is a discussion on how things are practiced TODAY!

If we want to assign credits to how we got here, why stop there and not go all the way back to our ancestors in the caves?!

Furthermore, it is misleading to think that our system of systematic thought and thinking as solely the realm of "philosophy", considering that the separation between philosophy and science did not exist then. I could easily argue that what you claim to be a philosophical product, is really a science product, and that those people were, in fact, doing science.

The fact remains that the training of scientists today generally requires ZERO formal training in the subject of philosophy. You may want to argue that science has benefited from the historical progress in philosophical thought. But that still doesn't change that we don't really care what is going on in the area of philosophy right now to do science. I could also argue that philosophy has benefited from scientific thought - Evolution, Relativity and Quantum Mechanics certainly came out of science FIRST.

Is this still an utterly incoherent argument?

Zz.

Scott Weingart said...

Of course science has influenced philosophy, and philosophy has influenced science. Physicists, these days, also use statistics. If statisticians stopped developing new statistical techniques, physicists today would not be the poorer... but current or future statistical methods may eventually become useful to future physicists, who themselves might learn those statistical techniques from physicists rather than the statisticians who developed them.

I'm not sure where this argument is going; if your only claim is that physicists don't care about philosophy, even though many physical discoveries and methods were influenced by philosophy, and that philosophers could stop working now and physicists wouldn't care, sure, I guess that's right. But so what? I could make the same argument about statistics, but I doubt any physicist would then say statisticians could stop working now, because their work has no bearing on physics..

Martin said...

Ah OK I was under a misapprehension about the degree of our disagreement. But I wasn't trying to make a historical argument, merely trying to illustrate that the science of any age is inseparable from the system of thought by which it is conceived. That's as true today as it's ever been.

Where I think we'd both agree is that the roles of the philosopher and the scientist have diverged in the last five hundred years. Newton and Descartes were philosophers *and* scientists. For them, considerations of 'method' - by which I mean methods of reasoning - were an important adjunct to experiment. Newton's rules of reasoning appear in the Principia, which is therefore a work of philosophy as well as of science.

Today the situation is rather different - there are few philosopher-scientists. But different scientific paradigms require different mental methods, and should today's scientific paradigm be succeeded by another, even more subtle (as the most cursory examination of the history of science suggests it will), those methods of reasoning will need reexamining once more.

But I'd just like to hammer one point home again: the methods of reasoning that underpin contemporary science are new (demonstrably, derived in the last 80 years), they are not immutable, and a scientist who thinks that the 'scientific method' has been perfected is naive indeed. Science and philosophy will be in dialogue for a long time yet.

ZapperZ said...

But to claim what I consider to be an INNATE ability of humans as being the product of philosophical thought is also giving philosophy way too much credit.

I have many friends who are not scientists. They were never trained as one. Yet, some of them approach things and do things systematically, very similar to what I would do as a scientist. Now, just because they do the same thing as I do, do I make the claim that they should credit scientists for the way they are thinking and approaching things? That would be absurd!

I would put it to you that many of the methodology that are used by scientists came out of an innate skill and ability that one either was born with (babies have been shown to understand science - see http://www.infozine.com/news/stories/op/storiesView/sid/50532/) or a skill that one picks up along the way as one works in this field! Just because what we do somehow has been studied by other great philosophers does not automatically mean that they came FROM them. Philosophy does not have a monopoly to claim all logic and analytical ability to be their derivative. This is not Al Gore taking credit for the internet!

Zz.

Martin said...

>>>I would put it to you that many of the methodology that are used by scientists came out of an innate skill and ability that one either was born with

Your assertion that scientific methodology is 'innate' is startlingly naive. It also suggests to me that you'd benefit from looking in more detail at the history of science, because it would show you that many of the tenets of contemporary experimental method that you take for granted are a) recently derived; and b) likely to change as the state of our knowledge does.


For instance, Aristotelian scientists (and there were two thousand years of them) privileged reasoning above experiment. If a physical demonstration failed to agree with the conclusions of logic, it was the 'real world' that was at fault, not their logic.

Take another example. Let's say you set up an experiment and repeat it six times. The results are pretty much identical each time. Until comparatively recently, the orthodox scientific view was that the results will be identical EVERY time you repeat the experiment. Today scientists have a far more nuanced view of induction (thanks to the nineteenth-century philosophers who argued this point) and they set a far higher bar on inductive conclusion.

A Newtonian scientist would have regarded his or her methods as 'innate', obvious and natural. But to you, five hundred years later, they're riddled with error. What feels natural to you (and me) is natural because we've been brought up in it. It doesn't mean it's inevitable and permanent.

>>or a skill that one picks up along the way as one works in this field!

But you 'pick up' these things from other people! And those other people have been educated. High school students learn about scientific hypotheses and the principle of falsification. That idea was first elucidated by Karl Popper, a philosopher! That one fact proves my whole case.

I agree with you that breakthroughs in science - and indeed the philosophy of science! - are far more likely to come from scientists than philosophers. But philosophy helps us to understand that all science rests on a massive and constantly-shifting edifice of human reasoning. There are huge dangers in assuming that science has reached some zenith of immutable and unimpeachable perfection in the way it treats experiment. That simply isn't true.

ZapperZ said...

Actually, I did study philosophy and history of physics quite a bit. Dan Siegel, who was a Professor at UW-Madison (and who was a physicist in a previous career) was my mentor. So I'm not claiming any of these out of ignorance.

But you've ignored completely my "experimental" point. Do you consider someone who isn't trained as a scientist, who follow the same scientific methodology, to actually learned such behavior through what philosophers have developed? There are many studies that will falsify that idea. Leon Ledermann, in fact, is pushing for physics to be taught very early in a child's education because he wants to tap into a child's curiosity and ability to learn things about the world around them. This is what I call an innate ability, and I'm NOT the only one! You learn about the world around you before you learn about learning!

Or are you saying that THAT too came out of those classical philosophers? Then maybe I should call them "God"?

Zz.

Scott Weingart said...

Martin's points are all fantastic, thank you for making them.

An innate curiosity is not equivalent to an innate understanding of the modern scientific method; that is something philosophers and scientists have cultivated carefully over quite some time.

It is terribly anthropocentric to assume that the human brain evolved in just such a way to understand the universe and the nature of learning perfectly. Atop the natural and "innate" is an architecture of millenia of learning and experience.

There is a reason people a thousand years ago, neither "scientist" nor "average joe," did not practice science as we know it today. We are born into a society that values certain ideas and methods above others, built by a complex web of scientists, philosophers, and others who all, themselves, drew (implicitly or explicitly) from one another.

Not everything from philosophy finds its way into a scientist's head, and what does might take a circuitous path of generations. But just as science inevitably influences philosophy, the same is true in the reverse, as ample examples in this thread have shown.

Martin said...

>>Do you consider someone who isn't trained as a scientist, who follow the same scientific methodology, to actually learned such behavior through what philosophers have developed?

'the same scientific methodology' is really too vague a phrase for me to answer this. But methods of reasoning are inherent to an entire school of education - an Italian humanist scholar of the humanities had an entirely separate logical approach from a nineteenth-century scholar, for instance.

>>There are many studies that will falsify that idea. Leon Ledermann, in fact, is pushing for physics to be taught very early in a child's education because he wants to tap into a child's curiosity and ability to learn things about the world around them. This is what I call an innate ability, and I'm NOT the only one! You learn about the world around you before you learn about learning!

You're talking about something completely different. Yes, a child has an instinctive grasp of how the world works - for instance an innate ability to predict what happens next if you drop a ball - and that is what Leon Ledermann and others have observed. This ability to grasp basic mechanics is light years from 'scientific method', which is to do with things like what constitutes a reasonable assumption, how to make and test a hypothesis and when to decide something has been 'proved'. Without education, a child is a pre-Aristotelian in his or her grasp of enquiry, and without education (or genius) will remain in this unsophisticated state.

ZapperZ said...

Ah, then that is where we differ, and that is why Feynman made that statement! The fact that there are many who are not trained either as philosophers or as scientists, and still managed to draw up accurate pattern of how things work is my proof that such ability is innate and requires no directions from such scholarly studies. Our ancestors understood how certain things are reproducible simply by looking at what worked, and what didn't. They didn't need a bunch of philosophers or scientists to tell them what to do.

And I think you missed the whole point of what Ledermann is doing. The whole idea here is that a child's NATURAL curiosity and ability is sufficient for him/her to draw up his/her own conclusion about things around him/her. This is without us having to teach the child how to learn things, or what METHOD to be used to learn such things. We can guide them, but they certainly do not need to be taught how to learn. This again proves my point that what these philosophers have studied are not something they invented! They may have clearly defined the steps and the what and whys, but they did not come up with it, the same way birds need not be told what they should and should not be doing.

Zz.

cephalopoda said...

But what about the forest boy of Borneo?

ZapperZ said...

What about him? He wants credit for the Scientific Method as well?

Zz.

cephalopoda said...

Well, what were the extent of his natural, innate abilities?