Editing Philosophy of science (section)

==History==
{{See also|History of scientific method|History of science|History of philosophy}}

===Pre-modern===
The origins of philosophy of science trace back to [[Plato]] and [[Aristotle]],<ref>
[[Aristotle]], "[[Prior Analytics]]", Hugh Tredennick (trans.), pp. 181–531 in ''Aristotle, Volume&nbsp;1'', [[Loeb Classical Library]], William Heinemann, London, 1938.
</ref> who distinguished the forms of approximate and exact reasoning, set out the threefold scheme of [[abductive reasoning|abductive]], [[deductive reasoning|deductive]], and [[inductive reasoning|inductive]] inference, and also analyzed reasoning by [[analogy]]. The eleventh century Arab polymath [[Ibn al-Haytham]] (known in Latin as [[Alhazen]]) conducted his research in optics by way of controlled experimental testing and applied [[geometry]], especially in his investigations into the images resulting from the [[Catoptrics|reflection]] and [[Dioptrics|refraction]] of light. [[Roger Bacon]] (1214–1294), an English thinker and experimenter heavily influenced by al-Haytham, is recognized by many to be the father of modern scientific method.<ref>{{cite book|last1=Lindberg|first1=David C.|title=Science in the Middle Ages|date=1980|publisher=University of Chicago Press|isbn=978-0-226-48233-0|pages=350–351|url=https://books.google.com/books?id=lOCriv4rSCUC&q=alhazen+philosophy+of+science&pg=PA351}}</ref> His view that mathematics was essential to a correct understanding of natural philosophy is considered to have been 400 years ahead of its time.<ref name="First Scientist, Clegg">{{cite book|last = Clegg|first= Brian|title=The First Scientist: A Life of Roger Bacon|publisher= Da Capo Press|date=2004|page= 2|isbn = 978-0786713585}}</ref>

===Modern===
[[File:Francis Bacon statue, Gray's Inn.jpg|thumb|Francis Bacon's statue at [[Gray's Inn]], South Square, London|350x350px]]
[[File:Hierarchy of the Sciences - diagram.svg|thumb|left|Theory of Science by Auguste Comte]]
[[Francis Bacon]] (no direct relation to [[Roger Bacon]], who lived 300 years earlier) was a seminal figure in philosophy of science at the time of the [[Scientific revolution|Scientific Revolution]]. In his work ''[[Novum Organum]]'' (1620){{mdash}}an allusion to Aristotle's ''[[Organon]]''{{mdash}}Bacon outlined a new [[system of logic]] to improve upon the old philosophical process of [[syllogism]]. Bacon's method relied on experimental ''histories'' to eliminate alternative theories.<ref>[[Francis Bacon (philosopher)|Bacon, Francis]] ''[[Novum Organum]] (The New Organon)'', 1620. Bacon's work described many of the accepted principles, underscoring the importance of empirical results, data gathering and experiment.  ''Encyclopædia Britannica'' (1911), "[[s:1911 Encyclopædia Britannica/Bacon, Francis|Bacon, Francis]]" states: [In Novum Organum, we ] "proceed to apply what is perhaps the most valuable part of the Baconian method, the process of exclusion or rejection. This elimination of the non-essential,&nbsp;..., is the most important of Bacon's contributions to the logic of induction, and that in which, as he repeatedly says, his method differs from all previous philosophies."</ref> In 1637, [[René Descartes]] established a new framework for grounding scientific knowledge in his treatise, ''[[Discourse on Method]]'', advocating the central role of [[rationalism|reason]] as opposed to sensory experience. By contrast, in 1713, the 2nd edition of [[Isaac Newton]]'s ''[[Philosophiae Naturalis Principia Mathematica]]'' argued that "...&nbsp;hypotheses&nbsp;... have no place in experimental philosophy. In this philosophy[,] propositions are deduced from the phenomena and rendered general by induction."<ref name="www.paricenter.com mullin02">{{cite web |url= http://www.paricenter.com/library/papers/mullin02.php |title= The Impact of Newton's Principia on the Philosophy of Science |last1= McMullin |first1= Ernan |website= paricenter.com |publisher= Pari Center for New Learning |access-date= 29 October 2015 |archive-url= https://web.archive.org/web/20151024002127/http://www.paricenter.com/library/papers/mullin02.php |archive-date= 24 October 2015 |url-status= dead }}</ref> This passage influenced a "later generation of philosophically-inclined readers to pronounce a ban on causal hypotheses in natural philosophy".<ref name="www.paricenter.com mullin02"/> In particular, later in the 18th century, [[David Hume]] would famously articulate [[skepticism]] about the ability of science to determine [[causality]] and gave a definitive formulation of the [[problem of induction]], though both theses would be contested by the end of the 18th century by Immanuel Kant in his [[Critique of Pure Reason]] and [[Metaphysical Foundations of Natural Science]]. In 19th century [[Auguste Comte]] made a major contribution to the theory of science. The 19th century writings of [[John Stuart Mill]] are also considered important in the formation of current conceptions of the scientific method, as well as anticipating later accounts of scientific explanation.<ref name="mill"> {{cite web
|url=http://plato.stanford.edu/entries/mill/#SciMet
|title=John Stuart Mill (Stanford Encyclopedia of Philosophy)
|publisher=plato.stanford.edu
|access-date=2009-07-31
|archive-url=https://web.archive.org/web/20100106122801/http://plato.stanford.edu/entries/mill/#SciMet
|archive-date=2010-01-06
|url-status=live
}}</ref>

===Logical positivism===
{{Main|Logical positivism}}
[[Instrumentalism]]{{technical inline|date=May 2025}} became popular among physicists around the turn of the 20th century, after which logical positivism defined the field for several decades. Logical positivism accepts only testable statements as meaningful, rejects metaphysical interpretations, and embraces [[verificationism]] (a set of [[epistemology|theories of knowledge]] that combines [[logicism]], [[empiricism]], and [[linguistics]] to ground philosophy on a basis consistent with examples from the [[empirical sciences]]). Seeking to overhaul all of philosophy and convert it to a new ''scientific philosophy'',<ref name=Friedman-pxiv>Michael Friedman, ''Reconsidering Logical Positivism'' (New York: Cambridge University Press, 1999), [https://books.google.com/books?id=e9TjZc9wNUAC&pg=PR14 p. xiv] {{Webarchive|url=https://web.archive.org/web/20160628112455/https://books.google.com/books?id=e9TjZc9wNUAC&pg=PR14 |date=2016-06-28 }}.</ref> the [[Berlin Circle (philosophy)|Berlin Circle]] and the [[Vienna Circle]] propounded logical positivism in the late 1920s.

Interpreting [[Ludwig Wittgenstein]]'s early [[Tractatus Logico-Philosophicus|philosophy of language]], logical positivists identified a verifiability principle or criterion of cognitive meaningfulness. From [[Bertrand Russell]]'s logicism they sought reduction of mathematics to logic. They also embraced Russell's [[logical atomism]], [[Ernst Mach]]'s [[phenomenalism]]—whereby the mind knows only actual or potential sensory experience, which is the content of all sciences, whether physics or psychology—and [[Percy Bridgman]]'s [[operationalism]]. Thereby, only the ''verifiable'' was scientific and ''cognitively meaningful'', whereas the unverifiable was unscientific, cognitively meaningless "pseudostatements"—metaphysical, emotive, or such—not worthy of further review by philosophers, who were newly tasked to organize knowledge rather than develop new knowledge.{{cn|date=March 2025}}

Logical positivism is commonly portrayed as taking the extreme position that scientific language should never refer to anything unobservable—even the seemingly core notions of causality, mechanism, and principles—but that is an exaggeration. Talk of such unobservables could be allowed as metaphorical—direct observations viewed in the abstract—or at worst metaphysical or emotional. ''Theoretical laws'' would be reduced to ''empirical laws'', while ''theoretical terms'' would garner meaning from ''observational terms'' via ''correspondence rules''. Mathematics in physics would reduce to [[symbolic logic]] via logicism, while [[rational reconstruction]] would convert [[natural language|ordinary language]] into standardized equivalents, all networked and united by a [[logical syntax]]. A scientific theory would be stated with its method of verification, whereby a [[logical calculus]] or [[operationalism|empirical operation]] could verify its falsity or truth.{{cn|date=March 2025}}

In the late 1930s, logical positivists fled Germany and Austria for Britain and America. By then, many had replaced Mach's phenomenalism with [[Otto Neurath]]'s [[physicalism]], and [[Rudolf Carnap]] had sought to replace ''verification'' with simply ''confirmation''. With [[World War II]]'s close in 1945, logical positivism became milder, ''logical empiricism'', led largely by [[Carl Gustav Hempel|Carl Hempel]], in America, who expounded the [[covering law model]] of scientific explanation as a way of identifying the logical form of explanations without any reference to the suspect notion of "causation". The logical positivist movement became a major underpinning of [[analytic philosophy]],<ref name="autogenerated1">See [http://plato.stanford.edu/entries/vienna-circle "Vienna Circle"] {{Webarchive|url=https://web.archive.org/web/20150810041731/http://plato.stanford.edu/entries/vienna-circle |date=2015-08-10 }} in ''Stanford Encyclopedia of Philosophy''.</ref> and dominated [[Anglosphere]] philosophy, including philosophy of science, while influencing sciences, into the 1960s. Yet the movement failed to resolve its central problems,<ref name="Smith1986">{{cite book |first=L.D. |last=Smith |year=1986 |title=Behaviorism and Logical Positivism: A Reassessment of the Alliance |publisher=Stanford University Press |isbn=978-0-8047-1301-6 |lccn=85030366 |url=https://archive.org/details/behaviorismlogic0000smit |url-access=registration |page=[https://archive.org/details/behaviorismlogic0000smit/page/314 314] |quote=The secondary and historical literature on logical positivism affords substantial grounds for concluding that logical positivism failed to solve many of the central problems it generated for itself. Prominent among the unsolved problems was the failure to find an acceptable statement of the verifiability (later confirmability) criterion of meaningfulness. Until a competing tradition emerged (about the late 1950s), the problems of logical positivism continued to be attacked from within that tradition. But as the new tradition in the philosophy of science began to demonstrate its effectiveness—by dissolving and rephrasing old problems as well as by generating new ones—philosophers began to shift allegiances to the new tradition, even though that tradition has yet to receive a canonical formulation. |access-date=2016-01-27 }}</ref><ref name="Bunge1996">{{cite book |first=M.A. |last=Bunge |year=1996 |title=Finding Philosophy in Social Science |publisher=Yale University Press |isbn=978-0-300-06606-7 |lccn=lc96004399 |url=https://archive.org/details/findingphilosoph0000bung |url-access=registration |page=[https://archive.org/details/findingphilosoph0000bung/page/317 317] |quote=To conclude, logical positivism was progressive compared with the classical positivism of [[Ptolemy]], [[David Hume|Hume]], [[Jean le Rond d'Alembert|d'Alembert]], [[Auguste Comte|Comte]], [[John Stuart Mill]], and [[Ernst Mach]]. It was even more so by comparison with its contemporary rivals—[[neo-Thomism]], [[neo-Kantianism]], [[intuitionism]], dialectical materialism, phenomenology, and [[existentialism]]. However, neo-positivism failed dismally to give a faithful account of science, whether natural or social. It failed because it remained anchored to sense-data and to a phenomenalist metaphysics, overrated the power of induction and underrated that of hypothesis, and denounced realism and materialism as metaphysical nonsense. Although it has never been practiced consistently in the advanced natural sciences and has been criticized by many philosophers, notably Popper (1959 [1935], 1963), logical positivism remains the tacit philosophy of many scientists. Regrettably, the anti-positivism fashionable in the metatheory of social science is often nothing but an excuse for sloppiness and wild speculation. |access-date=2016-01-27 }}</ref><ref>{{cite web|url=http://www.drury.edu/ess/philsci/popper.html |title=Popper, Falsifiability, and the Failure of Positivism |date=7 August 2000 |access-date=7 January 2014 |quote=The upshot is that the positivists seem caught between insisting on the V.C. [Verifiability Criterion]—but for no defensible reason—or admitting that the V.C. requires a background language, etc., which opens the door to relativism, etc. In light of this dilemma, many folk—especially following Popper's "last-ditch" effort to "save" empiricism/positivism/realism with the falsifiability criterion—have agreed that positivism is a dead-end. |url-status=dead |archive-url=https://web.archive.org/web/20140107230818/http://www.drury.edu/ess/philsci/popper.html |archive-date=January 7, 2014 }}</ref> and its doctrines were increasingly assaulted. Nevertheless, it brought about the establishment of philosophy of science as a distinct subdiscipline of philosophy, with Carl Hempel playing a key role.<ref name=Friedman-pxii>Friedman, ''Reconsidering Logical Positivism'' (Cambridge U P, 1999), [https://books.google.com/books?id=e9TjZc9wNUAC&pg=PR12 p. xii] {{Webarchive|url=https://web.archive.org/web/20160628220109/https://books.google.com/books?id=e9TjZc9wNUAC&pg=PR12 |date=2016-06-28 }}.</ref>

[[File:Epicycle and deferent.svg|thumb|For [[Thomas Kuhn|Kuhn]], the addition of [[Deferent and epicycle|epicycles]] in Ptolemaic astronomy was "normal science" within a paradigm, whereas the [[Copernican Revolution]] was a paradigm shift.|261x261px]]

===Thomas Kuhn===
{{Main|The Structure of Scientific Revolutions}}

In the 1962 book ''[[The Structure of Scientific Revolutions]]'', [[Thomas Kuhn]] argued that the process of observation and evaluation takes place within a "paradigm", which he describes as "universally recognized achievements that for a time provide model problems and solutions to community of practitioners."<ref>{{cite book|
last=Kuhn|first=Thomas|title=[[The Structure of Scientific Revolutions]]|year=1972|edition = 2nd| publisher=The University of Chicago|isbn=0-226-45803-2}}, p. viii</ref> A paradigm implicitly identifies the objects and relations under study and suggests what experiments, observations or theoretical improvements need to be carried out to produce a useful result.<ref>Kuhn clarified that these are two related senses of "paradigm": (1) "the entire constellation of beliefs, values, techniques" and (2) "the set of puzzle-examples which, employed as models or examples, can replace explicit rules as a basis" and are used to illustrate the field for beginners. {{harvnb|Kuhn|1972|p=175}}</ref> He characterized [[normal science]] as the process of observation and "puzzle solving" which takes place within a paradigm, whereas [[revolutionary science]] occurs when one paradigm overtakes another in a [[paradigm shift]].<ref>{{cite journal
|url = http://plato.stanford.edu/archives/fall2013/entries/thomas-kuhn/
|title = Thomas Kuhn
|access-date = 2015-10-26
|last = Bird
|first = Alexander
|year = 2013
|website = Stanford Encyclopedia of Philosophy
|editor1-last = Zalta
|editor1-first = Edward N.
|archive-url = https://web.archive.org/web/20170713100633/https://plato.stanford.edu/archives/fall2013/entries/thomas-kuhn/
|archive-date = 2017-07-13
|url-status = live
}}</ref>

Kurn was a historian of science, and his ideas were inspired by the study of older paradigms that have been discarded, such as [[Aristotelian mechanics]] or [[aether theory]]. These had often been portrayed by historians as using "unscientific" methods or beliefs. But careful examination showed that they were no less "scientific" than modern paradigms. Both were based on valid evidence, both failed to answer every possible question.{{sfn|Kuhn|1972|p=1-7}} 

A paradigm shift occurred when a significant number of observational anomalies arose in the old paradigm and efforts to resolve them within the paradigm were unsuccessful. A new paradigm was available that handled the anomalies with less difficulty and yet still covered (most of) the previous results. Over a period of time, often as long as a generation, more practitioners began working within the new paradigm and eventually the old paradigm was abandoned.{{sfn|Kuhn|1972}} For Kuhn, acceptance or rejection of a paradigm is a social process as much as a logical process. 

Kuhn's position, however, is not one of [[relativism]]; he wrote "terms like 'subjective' and 'intuitive' cannot be applied to [paradigms]."{{sfn|Kuhn|1972|p=175}} Paradigms are grounded in objective, observable evidence, but our use of them is psychological and our acceptance of them is social.