What's wrong with this definition of science?

Great Debates
1

Science: the investigation of Nature by means of instrumental measurement.

To clarify some details before the discussion starts (hopefully):

I use the word investigation to indicate that science proceeds by interrogating “naturally” occurring objects, and to differentiate science from other, non-investigative fields that also rely on instrumental measurement, such as engineering and house-building.

I use the word Nature in the classical sense, to indicate the natural world, the universe, Reality, etc.

I use the phrase instrumental measurement to indicate that the natural sciences progress by means of this methodology; that is, that they are primarily concerned with measuring the universe against a conventionally agreed-upon standard.

Obviously, I’m using the words “science” and “natural science” interchangeably here, in keeping with standard American English usage.

My definition does not concern itself with the cognitive contents of science, because I suspect that one cannot successfully locate “scientificity” as a characteristic of any given linguistic statement (which is also, IMHO, where Popper went wrong). Thus, any attempt to define “science,” or demarcate it from other forms of knowledge production, must concentrate on methods: that is to say, on what it is, precisely, that scientists do.

2

What’s wrong with the standard definition?

“The observation, identification, description, experimental investigation, and theoretical explanation of phenomena.”

3

Well, you’re also leaving out the concepts of “reproduceability” and “consistency”. It’s not so much of what scientists do so much as what their observations are after they do it.

4

Reproducibility and consistency are the epistemological requirements of some outlooks on what we can know from science. They needn’t be in the definition of science itself.

5

I’m not comfortable with the inclusion of “instrumental measurement” in your definition because it is not always a requirement for scientific endeavors.

For instance, the classification and naming of organisms as families, genera, species, etc. (taxonomy) does not necessarily depend on instrumental measurement or any kind of measurement in the physical/mathematical sense, only on systematic “observation, identification and description” as indicated by the standard definition. You may choose to view this as a kind of measurement, but surely not as instrumental measurement.

6

Science is a method, rather than a goal. Although most scientists do have goals, in many cases that goal is academic recognition, tenure, or monetary reward. Those goals do not necessarily mean that what they do is not science, that depends on their method.

Astrologers measure natural phenomena, and examine the relationships of those measurments with other natural phenomena. Phrenologists do, as well, and numerologists create mathematical models of reality. But what they do is not science, since they do not limit themselves to methods that meet the standards of science. A single sentence definition of Science is going to be hard to find, if you wish to have it include all the hallmarks of the method, and exclude the philosophies, theologies, arts, and sports which deal also with Man’s hunger to know the world.

Science is the exposition of natural phenomena, by the repeated, and recorded examination of quantified characteristics of those phenomena, by multiple examiners, in multiple iterations of documented methods, and the establishment of rigorously tested theoretical models to explain the mechanisms which cause those phenomena, and are subsequently caused by them.

What differentiates poor, mediocre, and great science is not the genius of the theoreticians proposing the theories, but the meticulousness of the observers in making, recording, and verifying the original data. Kepler was only able to make his fundamental discovery of the motions of the planets because he believed that the half minute arc difference in the values he observed in his calculations were not simply errors made by Tycho Brahe during the long years when he made his original measurments of observation. The button sorters are scientists, and without them, theories are simply daydreams in mathematical symbols.

Tris

“Here kitty, kitty, kitty.” ~ Erwin Schroedinger ~

7

I don’t see, Jerevan, why taxonomy (as you describe it) is a science and stocking shelves (done at supermarkets) isn’t.

Tris, I think the OP presents a good definition. I think instrumentation is a key to science. The OP did not offer how we distinguish good science from bad science, merely how one could classify all science.

Lib, I do not have to meet all those requirements to do science, do I? I cannot simply observe protien binding in different media at different pH levels and call it science? Must I explain why things are the way they are? Well, let us say, then, that the “and” is not a logical “AND”, then, so that we may take those different parts seperately from the rest.[ul][li]Observation of phenomena: looking out my window. Science![/li][li]Identification: that is, most assuredly, a turlingdrome. Science![/li][li]Description: turlingdromes resemble a cross between wheat and reed grass. Science![/li][li]Experimental investigation: They are flammable. Science![/li]Theoretical Explanation: turlingdromes are the result of crossbreeding wheat and reed grass, the conditions for which could only be found in animal dung. Science![/ul] Well, ok, maybe a little glib, there. But I think the OP has a good grasp of what we intuitively mean when we say “science”.

8

I think your definition is too limiting. Check out Karl Popper and David Deutsch for good explorations of the epistomology of science.

I think the Popper definition can be boiled down to:

Science is the creation of explanatory theories that are open to falsification.

Theory creation and falsification are both “doing science”. The theoreticians and the experimentalists are both working toward the same goal, just from different directions.

(Note that instrumental measurement is only one way to falsify theories. Logical consistency is another. Deutch adds the idea of explantory power to the mix – given competing non-falsified theories the preferable one is the one that provides the broadest explanation of the described phenomena.)

Putting the focus on “falsification” instead of on “experimentation” allows us to include activities that are clearly science (archaeology, for example, or astronomy) but are not experimental the way physics is. Coupled with Deutch’s ideas about explanatory power it also allows Popperian epistomology to be applied to fields outside of science.

(For example, my wife is a music historian. She uses the Popperian approach in her analysis of manuscript sources. It’s a general fomula for building knowledge, not just a way of doing science.)

9

I thought Popper felt that we could never gain inductive knowledge from experimentation, though.

10

As I mentioned, astrologers use instruments to measure natural phenomena. This entirely fulfills the definition presented. Do you consider astrology a science? I find the definition lacking for that reason. I don’t claim my own has no holes in it either. That is a problem often encountered when assessing the relative reliability of scientific research presented as fact. What is science, and how do you tell it apart from philosophy, artistry, or flim-flam?

Fairly often, pronouncements from academicians are presented to the public as scientific fact. Authoritarian opinions are revered as science, when in fact they are simply opinions. The objections or support of an experienced investigator are certainly worthy of some higher level of respect in the area of knowledge where he has specific research experience, but a Nobel Laureate in Physics who believes that massive doses of Vitamin C will cure the common cold is just another joe with an opinion. Science is all in the details of the method. Experiment design is critical, and precision in measurement and recording are quintessentially what make science a useful discipline. Without falsifiable prediction and testing, we have no means of establishing the level of internal consistency that has made that body of knowledge so reliable that it is perceived by so many as Science Itself.

While casual conversation will inevitably include such concepts as “body of knowledge” and “society of credentialed practitioners,” these elements cannot be included in the definition of science to scientists themselves, or it becomes another form of shaman arcana, meant to exclude the unanointed from recognition.

Tris

11

Granted, I know nothing about epistemology, but I think that reproducibility should be included in the definition of science. If you take the observations of UFOs for example, those observations are generally not reproducible and are therefore not considered to be real science by most scientists. They’re just one-time observations that provide no evidence for any kind of theory. Unless UFOology is just bad science, in which case I’m wrong.
I also think that science isn’t just about observation with instruments. When I do science in the laboratory, I don’t just observe the change of one variable with respect to another, I also try to see what that tells me about the nature of the system. The second part is just as important to science as the first. It may be true that what I deduce from an experiment is either wrong or based on prior observations and theories, which I suppose is methodology. That doesn’t mean that I can’t deduce(or induce?) anything.
(Hope I’m being clear…I sound confusing to myself.:))

12

I’m also a big fan of Popper and Deutch’s definition as given by Pochacco. As he (I assume) stated, it allows sciences like sociology, archaeology, and psychology to be counted as science, even though they are kinda different from physics and chemistry. Moreover, it disallows pseudosciences such as creation science and astrology, on the basis that their theories are not open to falsification.

I’d also like to add a third criterion to this: predictive power. To my mind, the most important aspect of science is what its theories say about data not yet gathered. Granted, not every scientific field has theories with equal predictive power (physics vs. archaeology, for instance), but all of them do have some.

13

Science is not the creation of theories that are open for falsification, for if they were, most scientists aren’t scientists. Most people don’t sit around all day theorizing, they’re collecting data and handing it off to someone else. That’s not science? In what world?

The more I hear about Popper the more I disagree with him. :mad:

Tris, I think that astrology is a science, yes. I think it is bad science, and that it cannot , by design, reveal what I would consider factual information, but I don’t see any reason to consider it not a science. The behaviors are remarkably similar. Numerologists, too. The whole shebang. Their work is sloppy and, I feel, largely useless, but I think it can be called science. Yep yep.

14

Eris,

On what basis would you include or exclude the following from a public school curriculum, or a publicly funded research program?

The approach of the Aquarian Age

Methods of sequestration for compounds containing Uranium.
The characteristics of superconductors in the 70 Kelvin temperatures.

God’s vilification of the serpent, as evidenced by the death of the so called dinosaurs.

I don’t think all of these are science. I don’t think science means thinking about stuff, and coming up will reasons for things.

I don’t insist that those which are not science be denigrated, forbidden, or excluded from my social matrix, but I feel that it is an intellectual disservice to group them together, and claim that they are all science.

In addition, as I am sure you and many others on the board know, I do not automatically grant supremacy to the view of science in all matters. I have faith in God, which in my opinion is an inappropriate venue for examination by the method of science. I don’t try to present my beliefs in the Lord as matters of science, although I do believe them to be facts, as real and as inevitable as gravity, or Quantum fluctuations in supersymetrical vacuum. (maybe more, who knows? :))

Neither do I find inherent fault in those who consult their astrologers before making life decisions, or examine the numerological portents of their daily experiences. They to are operating outside of the specific confines I believe are matters of science. If they pretend that their discipline is scientific, I will apply the rules, and requirements of science to their claims. Otherwise, I am not offended, or interested in the phenomena.

Tris.

“The road to truth is long, and lined the entire way with annoying bastards.” ~ Alexander Jablonski ~

15

Ditto, to the definition offered by Pochacco.

eris, OK, maybe taxonomy wasn’t the best example I could offer to make my point, and perhaps as in our earlier discussion about the uncertainty principle I am looking at the idea of an instrument too narrowly. Still, I don’t agree that instruments are – well, instrumental to Science. For instance, what about higher mathematics? Do differential equations or linear algebra require instrumental measurement?

One minor nitpick, though: if, in your response to Libertarian’s first post, your point was that an endeavor does not have to fulfill all of the requirements in his definition to be called “Science”, then I think we must allow that both taxonomy and grocery shelf-stocking are sciences. The latter even more so, since one stocks grocery shelves based on observations about what sells, and plans accordingly for future stockings. :wink:

ultrafilter, I think the criterion of falsification already implies that one uses theory to predict data not yet gathered, then one gathers the data and compares the observed to the predicted.

To eris’s latest post: you make an excellent point and I was wondering when it would arise, and who would raise it (long live Eris indeed). Astrologers (like my amateur-ish self) consider what they do to be a science, in the way psychology is considered a science, though we acknowledge that not everyone shares this perspective. So… I suggest that cannot ask the question “What is science?” without also (first?) asking, “Who gets to define it?”

16

**

Please go back and reread my original post. Collecting experimental data falls into the “falsification” part of my definition. I explicitly say as much.

Science is the creation of explanatory theories that are open to falsification.

Different people work on different parts of the problem. Some create explanatory theories. Some generate empirical data that can be used falsify theories. Both are doing science.

But if you’re creating theories that can’t be falsified, you’re not doing science. (Biblical literalists, for example.)

Similarly, if you’re making observations of the natural world that can’t be used for falsification then you’re also not doing science. (Looking out the window at the sunset, for example.)

The Popperian definition has the advantage of agreeing with our commonsense classification of who is a scientist and who isn’t. Astronomers are scientists, astrologers are not. Popper explanations why we’re justified in making this distinction.

That’s a pity, because he was one of the great thinkers of the 20th century.

Astrology and numerology resemble science only superficially. If I put on a white lab coat and wander around a hospital does that mean that I’m a “bad doctor”? Of course not! Dressing up like a doctor doesn’t make me a “bad doctor” – it makes me a “fake doctor”.

If you want to call astrology and numerology “fake science” or “pretend science” I won’t have any quarrel with you.

17

pochaco: Collecting data does not fall into the “falsification” part. Data is what would be used for falsification, but that was absent from that definition. The definition you gave dealt explicitly with the creation of theories that may be falsified. It said nothing about the act of falsification. Maybe you consdier it implicit, but I don’t see how.

Furthermore, many experiments aren’t done in order to falsify theories, they are done because the experimenters believe the theories are correct. This would also be out of the scope of your definition.

Now, you say that these individuals are taking part in science, well, ok I guess. I would argue that this isn’t how we use the term “science”, but if that is how you see it than what more can I say.

Well, he is on my reading list, just haven’t made it there, yet.

RE: the astrologers and fake doctors; again, I’m siding with the OP’s definition of science; I think it encapsulates what we think of when we say “science”, what “scientists” do, and why many seem to fall for “bad science” as “science”.

Jerevan:

Lib asked what was wrong with that definition. I responded that a requirement to meet all qualifications is too strict (and rules out some things we call science) and if we only consider them seperately than very strange things become science… things which don’t even resemble what we would normally call “science”. I make no claim that I considered any of those things science :wink:

For supporting Popper-cum-Pochaco’s definition, you might note that mathematical thoeries aren’t falsifiable and so don’t come under “science”. This is one reason why I disagree with that definition.

You said to ultrafilter: “I think the criterion of falsification already implies that one uses theory to predict data not yet gathered, then one gathers the data and compares the observed to the predicted.” I don’t see that. I see that a falsifiable theory must be such that if one had the instruments and ability to one could falsify the theory. How long after general relativity was offered did it take to perform experiments that could confirm or deny it with reasonable confidence? Should we say the Einstien did no science because his theory wasn’t falsifiable, or that he didn’t do science until, in hindsight, we confirmed his theory? Is he no better than an astrologer because his theory breaks down at the quantum level?

I find this Popper-whoever definition to be lacking as a definition of science. I find it to be tolerable as an epistemological requirement for good empirical science. At best.

Tris:
[ul][li]The approach of the Aquarian Age: is this something defined through use of tools, measurements, or the results of the use of tools or measurements? Then it is science. Is the Age of Aquarius meaningful? That’s another question. Is it considered an element of knowledge? That’s another question.[/li][li]Methods of sequestration for compounds containing Uranium.: An examination or explanation of these methods is not obviously science, though I can imagine scenarios where this would be.[/li][li]The characteristics of superconductors in the 70 Kelvin temperatures. * Same as above.[/li][]God’s vilification of the serpent, as evidenced by the death of the so called dinosaurs.: again, the context in which this is presented could make it go either way.[/ul]

18

For the record, I think the idea of “instrument” in the OP needs to be more clearly stated, and depending on how that goes the definition might need expanding a bit, too.

19

Well… partly correct. A lot of my scientific work involves problem-solving which to a large degree takes the form of eliminating possibilities – that is, first demonstrating that certain conclusions are not correct. While experiments are often constructed in an effort to prove a theory correct, at the same time they must be constructed in such a way that, if the theory is false/faulty, the results will demonstrate this. An experiment which does not allow the possible falsity of the theory to be demonstrated is not scientifically rigorous. A key point is that it only takes one negative result to disprove a theory; a positive result merely allows the theory to live until the next experiment.

OK. I thought that was your point, but I’m still getting accustomed to your style on the boards here. :wink:

Hmmm, I don’t feel qualified to comment on the place of falsifiability of mathematical theories. Someone with a stronger mathematics background will have to answer that. But I agree: if falsifiability does not apply to mathematical theories, then falsifiability is not a good criterion for defining science.

Yes, you are right. I ought to have said, “the criterion of falsification already implies that one could use theory to predict data not yet gathered…”, thereby removing the suggestion that one must be able to carry out an experiment. As you point out, there was a long gap between the development of Einstein’s relativistic theory and experiments which put it to the test, not for lack of experimental ideas, but lack of technology to carry out the experiments. This does not make his theories any less scientific, or make them retro-actively so after the experiments were carried out.

20

…or it means that certain areas of mathematics cannot be considered “science”.