The questions

Relation to truth· Truth, language and sciencedifficulty

Scientific method

How does science go about justifying its claims?

Science commands trust, but how exactly does it justify its claims: by proving them through experience, by exposing them to refutation, or by computing their probability? What we expect of scientific proof, and what separates science from mere belief, depends on it.

Proof by experienceRefutationThe calculus of probabilities
The poles, and who holds them

Proof by experience

5

Experience has the power to prove: a well-conducted observation, a rigorous experiment can establish a law or confirm a hypothesis. It is by accumulating positive evidence that science gains in certainty; experience is not merely a source of ideas, it is the tribunal that validates them.

Explore this position →
Associated figures: Bacon · Mill
Who stands here
Baconexplicit

The syllogism, Bacon objects, only links notions already received: it discovers nothing, it merely dresses up prejudice. To this logic of anticipations of nature he opposes a logic of interpretation: not to fly like the bee from experience alone, nor to spin like the spider from reason alone, but to climb step by step from facts to laws. His is therefore not the bare enumeration that generalizes from a few agreeing cases; it is an eliminative induction, drawing up tables of presence, absence, and degrees so as to rule out false causes methodically and isolate the form of the phenomenon. Science sets out from no preconceived system: it is built from the bottom up, by exclusion.

Novum Organum, I (aphorisms 19, 95, 105) · Novum Organum, II (tables and exclusion)

Millexplicit

The System of Logic codifies induction: it is the experimental methods (agreement, difference) that, from the observation of regularities, establish the laws of nature. Scientific knowledge is built from the particular toward the general.

A System of Logic, Book III

Empiricismexplicit

Since nothing is in the mind that was not first in the senses, the method can only be regressive then ascending: one starts from the observable given and rises to general laws by induction, never the reverse. Bacon erects this march into a rule, denouncing the hasty anticipation that imposes a priori principles on nature, and sets against it methodical induction, conducted through tables of absence and presence. This is the overturning of the rationalist model, which deduced the world from self-evident first truths: here axioms do not precede experience, they are its revisable outcome.

Locke, Essay; Hume, Enquiry; Bacon, Novum Organum

Show more (2)
The Enlightenmentexplicit

Hostile to a priori systems and gratuitous hypotheses, the Enlightenment makes observation and experiment the source of all physics: one rises from facts to laws by , in the manner of , whom the Encyclopédie claims as its patron. nonetheless radicalizes this programme by showing that induction itself is grounded not in reason but in habit. In parallel, the calculus of probabilities of Laplace and opens a quantitative path to uncertainty.

Bacon, Novum Organum · Condorcet, Essay on the Application of Analysis to the Probability of Majority Decisions

Charles Sanders Peirceinferable

None of the three poles captures Peirce exactly, for he added precisely the missing piece: , the inference that invents the hypothesis, upstream of the deduction that draws out its consequences and the that tests them. Science for him is this complete cycle, not one of its moments. He is nonetheless ranged on the inductivist side, since he remains an experimenter who grounds everything on the test of facts and makes induction the decisive trial; but he claims to defuse the flaw denounced by holding that repeated induction corrects itself in the long run. His training as a statistician also gives him a probabilist strand, attentive to frequencies and to error, that foreshadows Bayesianism without reducing to it. What makes science is not a single rule of inference but the long-run self-correction of a accountable to the real.

The Fixation of Belief (1877) · Collected Papers, V, 180-212 (on abduction)

Refutation

0

A theory is never proven, only refuted. A hypothesis is scientific if it exposes itself to an experiment that could disprove it; as long as it withstands them, it is corroborated, never established once and for all.

Explore this position →
Associated figures: Popper
Who stands here

No figure stands here, in this slice.

The calculus of probabilities

0

To know is to adjust degrees of belief. Each observation revises the probability of a hypothesis by Bayes' theorem: evidence does not prove, it makes more or less probable, starting from what one already believed.

Explore this position →
Associated figures: Ramsey · Carnap
Who stands here

No figure stands here, in this slice.

Youyou have not taken a stand on this axis yet.Locate yourself5 figures · 2 movements mapped
Problems

See the problem map in the Atlas →

  • Descriptive Thought experiment
    The black swan

    No number of white swans proves that all are white, but a single black swan is enough to belie it: how, then, does one establish a general law, by piling up cases, by hunting the counter-example, or by weighing the probabilities?

    a universal law
Going further

Sub-problems to deepen this axis's discussion

  • Descriptive
    What can evidence do?

    No accumulation of evidence proves a universal law; a single counter-example, on the other hand, seems enough to refute it: evidence would then count only by its power to belie, never to establish. But does it even truly refute? When a result contradicts a theory, it is never that theory alone that is at stake, but a whole bundle of auxiliary hypotheses, and one can always sacrifice one of them to save it. Prove, refute, or only make more or less probable: what, exactly, can evidence do?

    a piece of evidence, a test
  • Descriptive
    Measurement and its instruments

    An instrument does not record a bare fact: a thermometer presupposes a whole theory of heat, an apparatus assumes hypotheses about what it captures, and every calibration rests on other measurements, themselves fallible. The precision it displays may thus silently carry forward debatable theories, and exactness become a pretension rather than a fact. How far does a measurement establish the real, rather than already giving it form?

    the measuring instrument
  • Descriptive
    Are there bare facts?

    Science is supposed to start from the facts, neutral and given. But do we ever observe a pure fact? What we notice, what we take for a result, already depends on our expectations, our instruments, and the theory we have in mind: two scientists before the same sky do not see the same thing there. If observation is always oriented, what does science take its support from?

    observation
  • Descriptive
    What confirms a hypothesis?

    "All ravens are black" is equivalent to "everything non-black is a non-raven", which a green apple would then confirm, without one's having seen a single raven. And the same past observations equally support incompatible predictions. What, exactly, does a piece of evidence confirm, and by how much?

    the link between evidence and hypothesis
  • Descriptive
    Is there a method at all?

    What if there were no universal rule of sound scientific reasoning? Feyerabend argues that the great discoveries often broke the canons of method, and that in science "anything goes": the logic reconstructed after the fact is not what actually advances knowledge.

    the methodological rule itself

Related problems to broaden toward other axes

  • Descriptive
    From observed to unobserved

    A general law extends to all cases what has been established of only a few: that the sun has always risen does not prove it will rise tomorrow, for nothing guarantees that the unobserved will behave like the observed, nor the future like the past. No accumulation of evidence fills this gap: it is the challenge each method must meet in its own way.

    the inference from observed to unobserved
  • Descriptive
    Science or pseudo-science?

    What separates a science from a doctrine that puts on its airs, astrology, a theory that nothing could belie? The question of demarcation extends that of method.

    a doctrine claiming to be scientific
  • Descriptive
    One method, or many?

    Does the method that tests the laws of nature hold for history, society, the mind? To explain by causes, as in physics, or to understand meanings from within, as Dilthey and Weber would have it: do the human sciences have their own logic, or does a single method govern all knowledge?

    the sciences of mind and society
  • Descriptive
    Science as it is actually done

    Does the philosophers' logic describe what laboratories actually do, the discussion, the consensus, peer review, the building of models and simulations? Or is the real method first a social practice, which the logical reconstruction idealizes after the fact?

    the actual work of laboratories
  • Descriptive
    Will tomorrow resemble yesterday?

    The sun has always risen, but what guarantees it will rise tomorrow? No accumulation of observations proves a law: does our trust in the future rest on reason, or on habit alone?

    a law drawn from experienceRelation to certainty
  • Descriptive
    What makes something a science?

    What separates a science from a belief that puts on its appearances, astrology or a doctrine that nothing could disprove? Its capacity to expose itself to the verdict of the facts, or is another criterion needed, if a sharp boundary exists at all?

    a theoryPlace of science

All problems →

Quotations