Science Without Method

By Neuroskeptic | August 5, 2011 8:50 am

Everyone knows that The Scientific Method is the key to doing science. No-one’s quite sure what it is, but they know it’s there, and it’s something rather special.


It’s not. When scientists sit down to work, we don’t use “the scientific method” to make discoveries. We use microscopes, brain scanners, telescopes and particle detectors, all of which are just ways of looking at things. They’re special in terms of what they let you look at, but that’s it. Science is looking.

It’s true that in order to do good science, you need to be careful. You need to avoid falling into various traps that lead to misleading data and false conclusions. You could call the care taken over scientific observations “The Scientific Method”, and some people do, but that’s misleading, because none of it is specific to science.

One of the most important considerations in science is making make sure that you have a proper control condition. This sounds technical, but all it really means is that you need to make sure that you really are looking at what you set out to observe.

To discover the effect of a drug on people, say, you just give them the drug and look to see what happens, using the appropriaye equipment. However, you need to compare this to an appropriate control, such as a placebo pill, because if you don’t, you’re not just seeing the effect of the drug, many other things as well, such as the placebo effect, the passage of time, random events.

In the same way, if you wanted to find out what happens when you push that little button on your TV remote, you wouldn’t mash five other buttons at the same time. To discover what was in the top drawer of your dresser, you’d look there, not in the bottom drawer.

That’s really all there is to it. It can be complicated to do this in practice, but the principle is that simple: you take care to look at what you’re interested in.

It’s said that part of the “Scientific Method” is forming hypotheses, or theories. Scientists do that, but so do we all, all the time. You might have a theory that your boss is an alcoholic, or that your husband is cheating on you, or that your car’s spark plug is bust.

You might call these ideas, notions, hunches, suspicions, thoughts, fears, but they’re still hypotheses about the world. Indeed, scientists often use those words too. One word is as good as another.

If your boss was an alcoholic, the way to prove it might be to somehow give him a breathalizer test after lunch, or sneak a peek at his credit card bill and see how much he spends on booze. That would be an observation to test your hypothesis, or in other words, an experiment (another formal word that scientists don’t always use).

That’s all science is. Looking at things carefully, getting ideas, and checking them out.

I said this in my last post, but it bears repeating: this is why most objections to, or concerns about, “science” or worse “modern science”, fail. Any given scientist, or any given scientific theory, may be wrong, just like anyone or anything else. Yet to say that “Science can’t” do something is saying that looking and thinking can’t do it. To blame “Science” for something is to blame the human mind.

Note: This post is a follow-up to Science Doesn’t Say, and the second in a three-part series.

CATEGORIZED UNDER: philosophy, science
  • http://www.blogger.com/profile/11279402169161555639 Michelle Dawson

    Since I've been a researcher, I think I've only said “scientific method” once.

    This was in the middle of legal proceedings. The hearings were arduous and went on and on. At one point I interrupted everything (I was representing myself) and said, loudly, “this is really making me appreciate the scientific method.”

    As least in that one context, I think what I said was accurate. Somewhere, it's on the record.

  • http://twitter.com/realcubiksrube cubik

    I think you are talking about the difference between hypothesis-oriented vs. discovery-oriented science, right?
    BTW… Paul Feyerabend has some really interesting things to say on the topic of scientific method you might like as well (if you're up for something provocative) for example in his book “Against Method”.
    :)

  • Anonymous

    I think the “scientific method” is simply honesty. For example, not faking results.

    It is no different from the method used by good historians.

  • Anonymous

    hi neuro. how much are your thoughts inspired by richard feynman?

    i feel these, say “scientific things”, the same way you do and i remember that for me it was kind of “iterrative” process to get there. being reasearcher confronted by lust for immediate success in my coleagues i still had doubts about what was and wasn`t scientific, how reaserch should and shouldn`t be done. it forced me to think and to find out what was (or looked to me to be) crucial and what was just a bla bla around. and then i got to feynman`s essays (“cargo cult science” and those in “the meaning of it all”) and thoughts(as quoted in mlodinow`s book “feynman`s rainbow) and i was amazed. it was all there, 100% agreement. since then i`ve red other descriptions but none of them more accurate than those of feynman. and yours is just so similar… :-)

  • omg

    you could say that about cooking, craft.. maybe it just comes down to regulation.

  • http://emmiscafe.wordpress.com/ Emmy

    Scientific method does sound vague; but I took it to mean a group of specific precautions one takes in order to not skew research. One example would be combing your methods for anything that would bias the data. As opposed to ignoring that aspect. Scientific method.

    Eugenie Scott in a series of videos taught everything one needs to know about facts, hypothesis, law and theory. They have specific meanings. From the video:

    Fact: a confirmed observation. Believe it or not, facts can be wrong.

    Hypothesis: a testable statement, for which at least some evidence has been gathered. A statement of relationship.

    Law: a descriptive generalization.

    Theory: highest construct in science. A body of work that gives us reliable data. Theories explain laws and facts. A logical construct.

  • http://www.blogger.com/profile/06815277098386812048 Jayarava

    What you say is no doubt true of working scientists. But when I learned science I learned a particular method of doing it that involved the traditional jargon and procedures.

    I would make the comparison between a scientist and a musician. A musician has to learn theory: scales, harmony, rhythm. And for the first few years they have to think in these terms. A working musician will have internalised all the basics and not have to think of them consciously – they know when they need to count bars, and when they can just feel the pulse and go with it. They know when a note from “outside the scale” will add colour used in passing. If the key changes they don't have to stop and think what the new notes are.

    Similarly a well trained scientist may not go through the process or use the words they learned in secondary school. But they can do that only because of the long training they've had, they've internalised a process.

    “Scientific method” is something that philosophers of science say more than scientists I think.

  • http://www.blogger.com/profile/06647064768789308157 Neuroskeptic

    Anonymous #1: Honesty is a big part of it, and as you say, that's true of all academia. But you also need to make sure that you're not fooling yourself, and this is where things like control conditions come in. As Carl Sagan said “Science is what we have learned about how to keep from fooling ourselves” although I think this is still slightly misleading since none of it is specific to science, except in the practical details.

  • http://www.blogger.com/profile/06647064768789308157 Neuroskeptic

    Jayarava: That's a good analogy.

  • omg

    When chefs sit down to work, we don't use “the cooking method” to make dough. We use microwaves, spatulas, tea towels and pestle grinders, all of which are just ways of cooking things. They're special in terms of what they let you cook, but that's it. Cooking is cooking.

  • http://www.blogger.com/profile/13670447320400400816 ramesam

    I seem to have a problem with the framing of the essay – well, the central point which it appears to be addressing to. Is it more inclined to comment on tech research (i.e. the research work regarding mostly apps developed on the basis of some fundamental science?) – for, Neuroskeptic says: “you need to make sure that you really are looking at what you set out to observe.”

    No, scientific method at fundamental basic research level (blue sky) is about “falsification”, “verification”, cross-checking and “validation.” Yes, it is kind of “unbiased” looking; but a looking at and for an “unknown”, not always what you expect to see. And it has a method. Be wary of the possible “errors” creeping at every point (framing the null hypothesis, observational errors, sampling errors, instrumental and laboratory errors, time drift in experimental results etc.); and also hazards in interpretation like confirmation bias, framing problem, endowment effect, bandwagon effect, backfire effect etc. That is what is referred to as “Scientific method”, IMHO.

  • Anonymous

    Sure, the “scientific method” they talk about in elementary school is oversimplified. It's arguably even wrong, because it ignores biases induced by the selection of hypotheses to be tested.

    However, if you guys were to follow some formal methods a bit more rigorously, and demand the rigorous validation of the method and its area of application maybe you wouldn't have quite so many problems with bad statistics, publication bias, bias in the selection of what gets funded in general, unrecognized and untested assumptions, “hot” area effects, general overinterpretation, etc, etc, etc. All of which appear to be rampant in the literature of even the “hard” natural sciences.

    It's HARD to get real knowledge, especially once you get beyond easy to control things like basic physics. Yes, you have to think all the time; that's part of validating that what you're doing makes sense in the context where you're doing it. But you also shouldn't discount the idea of following some rules, particularly about formal validation and verification.

    Sometimes simple observation just doesn't cut it.

  • Anonymous

    I think that you, as a scientist, overestimate the logical abilities of people who are not educated in using 'the scientific method'.

  • http://www.blogger.com/profile/06647064768789308157 Neuroskeptic

    ramesam: There are plenty of things you need to take care of, yes, and these can be difficult in practice.

    My point here is that they're not difficult to understand conceptually. Most boil down to making sure you're observing what you're aiming to observe, and not an artefact of one kind or another.

    As for cognitive biases such as confirmation bias, again, that's right. But it's nothing specific to science, that's a problem for everyone, it's a problem with the human mind in general.

    I'm not saying that there is no method to science – but it's not a scientific method. Nothing in it (except the practical details) are special to what we call science.

  • dr_popper

    This post and many of the comments on it so far typify precisely what is WRONG with modern science.

    None of us seem to remember why the scientific method is important (or what it is, apparently), or why theory-testing is crucial.

    Read Popper or Hume and you will understand why you will not discover anything just by 'looking'. The problem is in the inductive reasoning required to discern anything meaningful from simple observation. Observe all you want, even build a career around it, but you will not get close to any kind of “truth” in doing so.

    More of the history and philosophy of science should be taught in undergraduate courses. Psychology especially, where most of our research is on WEIRD participants (http://dx.doi.org/10.1017/S0140525X0999152X).

  • omg

    REAL LIFE should be taught in schools. Clearly they're brainwashing peeps into industry crackheads subservient to industrial production so an elite few can bask with that knowledge.

    A regulated industry has restrictions. Anyone can come up with genius methodologies and Earth shattering revelations, but what good is it when the common man lacks the attentional resources to deem it relevant?

    Maybe the point of a method is to provide insight into replication and generate solidarity for future solutions – regulated because you don't want ethically disturbing methods (like torture) to be known to society.

    Off the radar, like let's say war time experiments, I bet they weren't too gung-ho about method.

  • omg

    Like the Mad Hatter, no-one cares what he says but somehow it's relevant because he can throw such a cool tea party. He probably calculated the temperature in the tea, adopted pristine cutlery to keep the percolation precise. Does Alice notice? No ? She just drinks it? I mean it may've crossed her mind omg he's hot or omg he's so weird but that's probably as far as she and others can digest. If the human mind is one of Yahweh's finest creation, then we're probably prone to abusing it. I guess it constantly needs regulation so it doesn't end up abusing society. So I reckon having a method keep things in check.

  • http://www.blogger.com/profile/06647064768789308157 Neuroskeptic

    dr_popper: Forming and testing theories is important – though it's not specific to science.

    However science is about much more than theories.

    Maybe physics is all about theories. Chemistry you could argue that it is. But biology certainly isn't. A lot of biology consists of collecting data, in the context of a well-understood theoretical background – or “just looking” (albeit technically).

    Take the Human Genome Project. That is Science if anything is. But it didn't aim to test a theory. It aimed to map the human genome, and it did.

    There's a view of science in which that kind of thing is “just stamp collecting”, with real science being about general Laws of Nature. But that strikes me as silly. If the HGP isn't science, then your definition of science is too narrow.

  • dr_popper

    Watson and Crick may disagree with you on that one. They were theoretical biologists and discovered the structure of DNA having never set foot into a lab. Biological research at its best!

    The HGP is 'observational' science, yes, but genes were predicted to exist LONG before ever being observed. That is how we knew to look for genes or a genome in the first place (and then later map). Because this search would falsify or support the theory of genetics. Thus, theory predicted the existence of a phenomenon before actually observing it, just like black holes and, now, dark energy in physics.

    Forming and testing theories is what differentiates science from all other fields of human endevour and the weakness of this in psychology is in large part what makes it a 'soft science'.

    Respectfully, I think your definition of science is the narrower one. Just because a scientist makes observations does not make 'observing' science. Scientists also check emails and drive cars, but these things are not science.

  • http://www.blogger.com/profile/06647064768789308157 Neuroskeptic

    dr_popper: Watson and Crick certainly were theorists, but that's why I chose the example of the HGP :)

    It's true that the HGP was, implicitly, testing Watson and Crick's theory of DNA.

    But only implicitly. That wasn't the point of the project. The project was to map the genome.

    In the same way, Columbus's expedition was intended to explore India (and ended up exploring America).

    Now implicitly, it was also testing, say, the theory that wooden ships float on water.

    But not it any interesting sense. And I don't think the HGP was testing Watson & Crick in any meaningful sense.

  • dr_popper

    I think you misunderstood my reason for discussing Crick and Watson. It was only as a response to your comment: “Forming and testing theories is important – though it's not specific to science” and that theories are 'less important' in biology than physics. I was not suggesting that the HGP was in any way a test of Crick and Watson's theoretical work.

    This has been an interesting discussion and I appreciate you continuing with it. However, I am still not persuaded by your argument and still believe strongly that neuroscience and other behavioural sciences would benefit from greater focus on theory.

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Neuroskeptic

No brain. No gain.

About Neuroskeptic

Neuroskeptic is a British neuroscientist who takes a skeptical look at his own field, and beyond. His blog offers a look at the latest developments in neuroscience, psychiatry and psychology through a critical lens.

ADVERTISEMENT

See More

@Neuro_Skeptic on Twitter

ADVERTISEMENT
Collapse bottom bar
+

Login to your Account

X
E-mail address:
Password:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »