This thought has been on my mind for a long time. It was reinvigorated by this question.

The most cliche presentation we hear about what “science” is in everyday discourse is something like:

  • a flowchart-type diagram involving “making observations”, “formulating hypotheses”, “making experiments”
  • Some very common re-hashing of Karl Popper that it “has to be falsifiable”
  • Or, probably some focus on “testability”, or “empirical evidence” or “experimental evidence”

Now, on the one hand, if the key distinguishing feature of “science” is empirical knowledge, as opposed to, I don’t know, Descartes-type a priori rational knowledge, then I think this idea of “science” would encompass more than is commonly considered “science” (like, everyday first-person experiences are “empirical knowledge”, like going to an ice cream shop), and it would fail to address what is probably a key aspect of “science”, which is the rational component (ie, how much of “science” actually is about non-empirical analytical reasoning; building theories, mental models, etc.)

The question is of importance to me, since I have often felt that “science” has become a very effective rhetorical instrument in this period of history. I have often felt one of the best ways to immediately persuade someone of anything is to tell them that it’s “science”. It’s sort of a Trojan horse in which because they believe “science” is the antidote to all forms of dubious and irrational thinking, as long as they can become convinced something is “science”, they will believe almost anything, even if it is actually something lacking in epistemic soundness. (I don’t just mean cliche ideas of “pseudo-science” like New Age medicine, but rather, how cultural dynamics reinforce what ideas count as “science” vs “pseudo-science”, and those words alone are enough to steer people’s epistemic allegiances).

What is strange is that the word “science” comes from a Latin word meaning “to know”; “philosophy” comes from a Greek word meaning “wisdom”; “mathematics” comes from a Greek word meaning “to learn”; and I think “ethics” comes from the Greek “ethos”, which meant “character, disposition or personality” but originally came from “accustomed place” (like the habitat of say, horses), and “morality” also has a strikingly parallel etymology in Latin, coming from “mores” (i.e. customs, habits) (I think, anyway). The point being, a lot of terminology with a lot of concentrated meaning, if you try to peel back the layers and get down to some semantic primitives giving them strong definition and demarcation, are kind of empty. “Science” means (literally) nothing but “knowledge”. Are we conditioned through experience to associate “science” with particular techniques we happen to see in our society around us, like laboratories, microscopes, “double-blind placebo controlled studies”, “Ph.D.’s”, etc.? Does “science” reduce to nothing but epistemology, the analysis of the nature and conditions of “knowing” something?

The SEP features this paragraph which I like:

Among the activities often identified as characteristic of science are systematic observation and experimentation, inductive and deductive reasoning, and the formation and testing of hypotheses and theories. How these are carried out in detail can vary greatly, but characteristics like these have been looked to as a way of demarcating scientific activity from non-science, where only enterprises which employ some canonical form of scientific method or methods should be considered science (see also the entry on science and pseudo-science). Others have questioned whether there is anything like a fixed toolkit of methods which is common across science and only science. Some reject privileging one view of method as part of rejecting broader views about the nature of science, such as naturalism (Dupré 2004); some reject any restriction in principle (pluralism).

I’m gonna read through that article, but would definitely like some supportive discussion here to think things through and analyze different dimensions of the question in detail.

  • 1
    You are equivocating between science that obviously exist: it produced the knowledge that allow us to discuss here and now on this site, and scientific method that is quite hard to be defined. Commented Apr 6 at 15:49
  • 1
    @JuliusHamilton I consider the SEP article a good start. Notably its long list of further readings. - But your final prompt "[I] would definitely like some supportive discussion here to think things through and analyze different dimensions of the question in detail." seems much to general for this platform. What about studying some of the references and then coming back with a concrete, more focused question?
    – Jo Wehler
    Commented Apr 6 at 15:53
  • Understanding Reproducibility and Replicability: ncbi.nlm.nih.gov/books/NBK547546. There is a theory that the human brain generates knowledge via observation. So if a culture lives in the Arctic Circle there would be more words for weather conditions involving snow than for a culture living at the Equator. Hebrew language has hundreds of names for attributes of God. Mental disorders are invented categories by humans who map behavior as pathology but there are patterns expressed and experts can in theory replicate a diagnostic disorder. Group-think, pseudo-science, or science? Commented Apr 6 at 16:16
  • 1
    @Scott Rowe - Galileo's efforts evolved into empirical political efforts. NIST Guide to SI: nist.gov/pml/special-publication-811/…. But DSM is non-empirical: pubmed.ncbi.nlm.nih.gov/27650639 How Voting and Consensus Created the Diagnostic and Statistical Manual of Mental Disorders (DSM-III) - The paper concludes that while the APA represented DSM-III, and the return to descriptive psychiatry it inaugurated, as a triumph of empirically based decision-making, the evidence presented here fails to support that view. Commented Apr 6 at 22:14
  • 1
    @Scott Rowe - Out beyond ideas of wrongdoing and rightdoing, There is a field. I'll meet you there. When the soul lies down in that grass, The world is too full to talk about. Ideas, language, even the phrase each other Doesn't make any sense. poetrysociety.org/poetry-in-motion/… The Essential Rumi. Sigmund Freud had a love-hate relationship with his father and mother who no doubt found fault with him. He sexualized his pain. He found fault with his patients and took their money so he could sex his wife and conform to patriarchal norms! Commented Apr 6 at 23:03

4 Answers 4


Is there such a thing as “science”?


“Science” means (literally) nothing but “knowledge”.

as in some form of objective, immutable, infallible knowledge? of course not. as useful, coherent knowledge, that enables us to construct things, somewhat understand and predict natural phenomena, etc.? of course yes: just look at your phone/computer, or go to the nearest drugstore, or check the weather prediction

then I think this idea of “science” would encompass more than is commonly considered “science” (like, everyday first-person experiences are “empirical knowledge”, like going to an ice cream shop)

maybe we should take a more restrictive notion of 'empirical' here, instead of 'anything whatsoever that happens in the physical world', though i'm not sure which one exactly

since I have often felt that “science” has become a very effective rhetorical instrument in this period of history. [...]

the issue is very timely and interesting, and i strongly recommend reading my favourite answer in the site for more about it

Does “science” reduce to nothing but epistemology, the analysis of the nature and conditions of “knowing” something?

no, people often want some usefulness/applicability out of it too


There are several abstract objects called Science. Your bullet points capture the important ones:

  • a particular kind of method
  • the body of claims whose form allows them to be falsified by the above method
  • the body of data useful for the above method

These are different abstract objects, but their close relation makes it useful to refer to them by the same word.

There are some physical processes called Science. In particular:

  • the physical things that human practitioners do when they're practicing the above method.

...but also sometimes...

  • instructional texts which fall into a certain category
  • the physical representation, in computer files and paper documents, of the pertinent data and claims.

Then we have some other abstract objects and physical processes that exist which are called Science, that really shouldn't be, but I'm not the definition boss:

  • the body of claims made by certain high-prestige people
  • those people themselves

Science is not, however, an agent. To fit your title question, we can rephrase that as: "There is no such agent as Science."

"Science knows," "science has shown," "science believes," "science prefers," "according to science," and so on are either potentially misleading shorthand or reflect cognitive error. If the speaker knows that he is anthropomorphizing a method or a heterogeneous body of claims and data - hopefully to an audience that knows what he's doing and won't be misled - it's shorthand. If the speaker is doing that without realizing what's happening, it's cognitive error.

Likewise for Humanity, Society, The Free Market, The [Nationality] People, The Proletariat, etc.


Science originally meant an organised body of knowledge. This is why in the Islamic world people can say the Qu'ranic sciences and the science of Hadith.

This is backed up by the etymology of science which derives from the Latin scire which means to know. Further, we can think of it as justified knowledge. And we can break that further down to justified by scripture where the Qu'ramic and Hadith sciences lie and to justified by empirical means. This is where science in the usual sense of the word lie. There are other bifurcations where speculative science such as the multiverse and string theory lie.

  • 1
    You have to start somewhere, but you have to end up somewhere too. Writings that are only justified by other writings doesn't seem to arrive, as far as I can tell. The starting point might also be dubious. If people can't verify for themselves, you just have a bureaucracy.
    – Scott Rowe
    Commented Apr 6 at 23:11

Science shares most of its methodological principles with . . . nearly all humans, at least those which are rational.

Science also has a lot in common with religions, namely the formalisation of its beliefs for the purpose of teaching. Religions are not rational, science by definition is. One important consequence and distinguishing characteristic is that scientists can easily switch from one set of beliefs to another.

What distinguishes science from mere rational investigation of the world is that it is a systematic and collective effort sustained generation after generation. The consequence of this is that scientists have been able to learn quite a few things: They learned the value of a mathematical formalisation of their theories; they learned the value of observation and of specific instruments, including very costly ones; they learned the value of teaching their theories. They learned the value of cooperation.

Compared to science, religions are mostly not rational, philosophy is mostly not collective; mathematics is mostly not empirical.

Another crucial characteristic is that science has demonstrated its value to society. The consequence of this is that in many countries scientists are paid for doing science and get governments to finance their activity.

This seems good enough to me to say that science exists.

Not the answer you're looking for? Browse other questions tagged .