Philosophy of Natural Science

by Carl G. Hempel

This volume explores the logic and methodology of scientific inquiry rather than its substantive results.

Genres: Philosophy, Science, Nonfiction

128 pages, Paperback

First published January 1, 1966

Reviews

[Julia Tibblin · Rating 1 out of 5]

Good bye Hempel see you never

[Alatea · Rating 3 out of 5]

Very useful, but super hard to understand.

[J.D. Steens · Rating 2 out of 5]

The book was technical and dry. Most of it involved the hard sciences where definitions and data are precise.

Hempel clarifies that to describe phenomena and predict its behavior is not the same thing as explaining why it operates as it does. We have a “strong grounds for believing” that “galaxies recede from our local one at enormous speeds, yet [this] does not explain why.” Hempel also cautions that a scientific law “cannot be adequately defined as a true statement of universal form” without specifying the accompanying conditions (laws “are known to hold only approximately and with certain qualifications”). As an example of a qualified form pertains to “probabilistic laws” such as the possibility that measles might be contracted under certain conditions. Hempel does not discuss whether there are implications of quantum physics for the “philosophy of natural science.”

For humans, as natural phenomena studied scientifically, Hempel uses the “mind-matter psycho-physical problem” to specifically illustrate the division between reductionism (psychological phenomena that “can be reduced to those of biology, chemistry, and physics”) and something that exists beyond this. Hempel uses behaviorism as an example of reductionism that rejects “all reliance on methods such as introspection, which can be used only by the subject himself in a phenomenalistic exploration of his mental world; and it does not admit as psychological data any of the ‘private’ psychological phenomena – such as sensations, feelings, hopes, and fears – that introspective methods are said to reveal.” I am not clear whether he is supporting this kind of reductionism but, if he is, he misses a good part of who we are. It could be that Hempel is saying that there is no mind-matter problem (in the sense of an inherent division), but rather, that science has not identified how chemical-electrical signals are [‘mechanically’] transformed into the representation of reality and encaptualized as memory.

At the end of the book, Hempel refers to “methodological individualism” in the social sciences that explains individual behavior in terms of “individual psychology, biology, chemistry, and physics” but Hempel casts it outside of the realm of science. Or, rather, he says that this is a problem for the “philosophy of the social sciences,” not “the philosophy of natural science,” which is the scope (and title) of his book.

Hempel bumps into a problem with biology. It’s almost like he’s saying that biology is not a science. As it now stands, life phenomena are deemed to be “manifestations of underlying teleological agencies of a nonphysical kind, referred to as entelechies or vital forces.” Hempel cautions against the import of mysterious life forces that neither describe nor explain life phenomena with the precision necessary to be viewed properly as science. Biology remains, he says, a neo-vital discipline and science must continue to “persist in the search for basic physico-chemical theories of biological phenomena rather than resign himself to the view that the concepts and principles of physics and chemistry are powerless to give an adequate account of the phenomena of life.”

A clue to Hempel’s views on biology is given in this description: “Living systems…display a variety of striking features that seem to be distinctly purposive or teleological in character,” he writes. This “that seems to be” reference is interesting given that living beings are clearly purposeful. They do what needs to be done to survive. Acknowledging that biological processes are highly varied and not as neat and tidy as the non-life sciences, “it seems to be” a step too far to exclude life science from natural science. Is there a problem with the definition of natural science? Is hard science criteria appropriate for the “life sciences?” In this book, Hempel appropriates the “natural science” terminology when, perhaps, the book should have been titled the “philosophy of physical science” which, like biology, would be then a subset of “natural science.”

[Vasile Rotaru · Rating 4 out of 5]

This is a great, classical introduction to some traditional issues of the philosophy of natural science. There are, of course, better, more comprehensive and serious introductory books on this subject, but Hempel's could still be used as a starting point.

[tanvi · Rating 2 out of 5]

Read this for class. The best thing about it is that it's over.

[Carlos Sode · Rating 4 out of 5]

Obra estrella del filósofo alemán Carl Hempel, imprescindible para comprender su archiconocido modelo de cobertura legal desde el que se pretende fundamentar la explicacion cientifica. También incluye disertaciones acerca de la construcción y contratación de hipótesis así como de la formación de los conceptos científicos y la reducción de las teorías. Es interesante y amena de leer, pero la historia ya ha demostrado en reiteradas ocasiones las limitaciones del programa neopositivista.

[Marcel Santos · Rating 3.5 out of 5]

ENGLISH

Carl Hempel’s “The Philosophy of Natural Science” reads as a good, didactic overview of Philosophy of Science. There is a lot to learn from this work, even though it does not add any major innovations to this field of knowledge following the contributions of Popper, Kuhn, Lakatos, and Feyerabend.

The book is clear most of the times and starts out very much engagingly bringing the story of Ignaz Semmelweis as a good example of scientific investigation to solving a practical problem.

Between 1844 and 1848, Hungarian physician Semmelweis investigated an intriguing phenomenon occurring at the Vienna General Hospital. He noted that the First Maternity Division of the hospital was having a high rate of post birth mother deaths by puerperal fever, while Division two had a lower one. The situation was intriguing as even mothers who would give birth outside the hospital and checked in right after had lower rates of the disease, what indicated that the problem was inside the hospital. The doctor tried the most different measures to figure out what was going on. That included, for instance, a request for the priest who would give the extreme unction to stop ringing a bell.

After a series of attempts to figure out the cause of the problem, Semmelweis noted that the doctors were carrying out autopsies in the autopsy wing located at the First Division and then going directly to carry out birth deliveries without properly disinfecting their hands. He also witnessed a colleague (Kolletschka) dying with the same symptoms of puerperal fever after having his finger cut in an autopsy. These were times when microbes were not known yet, so Semmelweis attributed the disease and deaths to contamination by “cadaveric material”.

The solution was to make doctors carefully disinfect their hands with a chlorinated lime solution rather than just simply washing them inattentively before carrying out child delivery. This finally reduced the problem. He also noted that not only cadaveric matter could contaminate. What he called “putrid material from living beings” could also do it. In one occasion, Semmelweis and a team properly disinfected their hands and examined a woman in labor with a festering cervical cancer. However, they only superficially washed their hands when examined next twelve women in the same room. The result: eleven of them died of childbed fever.

Hempel uses that case as a good example of scientific inquiry. Semmelweis’s understanding of the factors contributing to the problem he faced was, by today’s standards, incomplete; yet the hypotheses he formulated and the tests he conducted helped him arrive at an effective solution to the problem.

Hempel recognizes that the mental activities to reach hypotheses and theories involve guesses and ingenuity — of which the Semmelweis case was rich. But guesses and ingenuity surely are a direct product of the knowledge already accumulated. A complete novice hardly reaches a scientific innovation. Hypotheses can be reached through “happy guesses” after observation of patterns and the use of induction. After seeing regularities, the explanation must reflect what one normally sees or experienced, but its testability is what makes the hypothesis scientific. Confirmatory testing does not provide certainty but can reinforce the hypothesis.

Hempel goes through the deductive-nomological inference, which is the classic structural reasoning model. It has the objective of explaining. In an explanation, the “explicandum” is the phenomenon to be explained. The “explicans” is what seeks to explain. The hypothetical-nomological model lays out as follows:

[law of nature]: All metals expand when heated;
[initial condition]: The cool sample of mercury was placed in hot water;
[initial condition]: Mercury is a metal;
[conclusion]: The sample of mercury expanded.

An important distinction, however, refers to regularities provoked by laws of nature and regularities by accident. Apparently there is no single piece of gold heavier than a hundred thousand kilos on Earth. Yet this is not due to some law of nature but to a coincidence. No law of nature prevents the production of a piece heavier than 100,000 kilos of gold by the fusion of two or more pieces into one. This may be because no one has tried to do this yet — likely because there is neither a reason nor an incentive to do so.

Unlike Popper, Hempel is no detractor of induction as a valid inference to scientific method. Induction is present in the analysis of data and it allows to formulate hypotheses — and it reveals itself on the basis of a probability. Just like this:

- Women who carry gene X mutation will probably develop cancer;
- Mary is a woman;
- Mary carries gene X mutation;
- Mary will probably develop cancer.

The structure resembles the deductive reasoning. However, unlike deduction — in which the conclusion is necessarily true provided the premises are true and the argument is valid —inductive reasoning establishes a hypothesis based on a series of regularities (e.g., most women develop cancer because this has occurred many times in the past) and arrives at a probabilistic conclusion. Also differently from deductive reasoning, the sentence “women who carry gene X mutation will probably develop cancer” does not stand for a law of nature, which would carry a general truthful rule (such as “gene X mutation will always develop cancer”), but a probability.

Another example of inductive reasoning, which admits flaw: in a man who carries one allele for brown eyes and another for blue, the probability for his child to be brown-eyed is theoretically 50%. This probability may not work thus if there are factors inherent to gene dynamics that could favor one or the other — and science could find this out by evolving its inquiries.

After reading the most classic authors of Philosophy of Science, it strongly feels like there is so little left in the subject matter to be added. The last book I read on the subject was Norwood Russel Hanson’s “Patterns of Discovery” (my review here: https://www.goodreads.com/review/show...), which seems to be the zest of the pot with the importance he gives to observation of data, the activities carried out prior to the formulation of a hypothesis, and his emphasis on Charles Sanders Peirce’s abductive/retroductive logical inference.

Hempel does not mention retroduction, as he focuses on the most known figures of deduction and induction. His explanation of them is among the clearest, though.

Just like Hanson’s book, Hempel’s “Philosophy of Natural Science” lacks a conclusion or at least a summary of the themes covered, which would certainly improve the entire reading experience.


PORTUGUÊS

O livro “The Philosophy of Natural Science”, de Carl Hempel, apresenta um bom e didático panorama sobre Filosofia da Ciência. Há muito que se aprender com essa obra, embora não acrescente grandes inovações a esse campo do conhecimento após as contribuições de Popper, Kuhn, Lakatos e Feyerabend.

A obra é clara na maior parte do tempo e começa de maneira bastante envolvente, apresentando a história de Ignaz Semmelweis como um bom exemplo de investigação científica voltada à solução de um problema prático.

Entre 1844 e 1848, o médico húngaro Semmelweis investigou um fenômeno intrigante que ocorria no Hospital Geral de Viena. Ele observou que a Primeira Divisão de Maternidade do hospital apresentava uma alta taxa de mortes maternas por febre puerperal após o parto, enquanto a Segunda Divisão apresentava uma taxa menor. A situação era intrigante, pois até mesmo as mulheres que davam à luz fora do hospital e se internavam logo depois apresentavam taxas menores da doença, o que indicava que o problema estava dentro do próprio hospital. O médico tentou as mais diversas medidas para descobrir o que estava acontecendo. Isso incluiu, por exemplo, um pedido para que o padre responsável por administrar a extrema-unção parasse de tocar um sino.

Após uma série de tentativas de identificar a causa do problema, Semmelweis observou que os médicos realizavam autópsias na ala de necrópsia localizada na Primeira Divisão e, em seguida, iam diretamente realizar partos sem desinfetar adequadamente as mãos. Ele também testemunhou a morte de um colega (Kolletschka), que apresentou os mesmos sintomas da febre puerperal após ter um dedo cortado durante uma autópsia. Naquela época, os microrganismos ainda não eram conhecidos, de modo que Semmelweis atribuiu a doença e as mortes a contaminação por “matéria cadavérica”.

A solução foi fazer com que os médicos desinfetassem cuidadosamente as mãos com uma solução de cal clorada, em vez de simplesmente lavá-las de maneira descuidada antes de realizar os partos. Isso finalmente reduziu o problema. Ele também observou que não era apenas a matéria cadavérica que podia causar contaminação. Aquilo que chamou de “matéria pútrida proveniente de seres vivos” também podia fazê-lo. Semmelweis e uma equipe desinfetaram adequadamente as mãos e examinaram uma mulher em trabalho de parto que apresentava um câncer cervical supurante. No entanto, ao examinar as doze mulheres seguintes na mesma sala, limitaram-se a lavar superficialmente as mãos. O resultado: onze delas morreram de febre puerperal.

Hempel utiliza esse caso como um bom exemplo de investigação científica. Semmelweis ainda chegou a uma compreensão insuficiente dos fatores que levavam ao problema que enfrentava, quando se compara ao que sabemos hoje; mas as hipóteses que formulou e os testes que realizou o ajudaram a encontrar uma solução eficaz para o problema.

Hempel reconhece que as atividades mentais envolvidas na elaboração de hipóteses e teorias incluem conjecturas e engenhosidade — características que não faltaram no caso de Semmelweis. No entanto, conjecturas e engenhosidade são certamente produtos diretos do conhecimento já acumulado. Dificilmente um completo iniciante chega a uma inovação científica. As hipóteses podem surgir de “palpites felizes”, após a observação de padrões e o uso da indução. Depois de identificar regularidades, a explicação deve refletir aquilo que normalmente se observa ou se experimenta, mas é a possibilidade de testá-la que torna científica uma hipótese. Os testes confirmatórios não proporcionam certeza, mas podem reforçar a hipótese.

Hempel examina a inferência dedutivo-nomológica, que constitui o modelo clássico de raciocínio estruturado. Seu objetivo é explicar. Em uma explicação, o “explicandum” é o fenômeno a ser explicado. O “explicans” é aquilo que busca explicá-lo. O modelo hipotético-nomológico pode ser apresentado da seguinte maneira:

[Lei da natureza]: Todos os metais se expandem quando aquecidos.
[Condição inicial]: A amostra fria de mercúrio foi colocada em água quente.
[Condição inicial]: O mercúrio é um metal.
[Conclusão]: A amostra de mercúrio se expandiu.

Uma distinção importante, contudo, diz respeito às regularidades provocadas por leis da natureza e àquelas que ocorrem por acaso. Aparentemente, não existe na Terra nenhuma peça de ouro que pese mais de cem mil quilos. Isso, porém, não se deve a alguma lei da natureza, mas a uma coincidência. Nenhuma lei da natureza impede a produção de uma peça com mais de 100 mil quilos pela fusão de duas ou mais peças de ouro em uma só. Isso pode ser porque ninguém tentou fazer isso até agora — provavelmente porque não há motivo nem incentivo para tanto.

Ao contrário de Popper, Hempel não rejeita a indução como uma inferência válida para o método científico. A indução está presente na análise de dados e permite formular hipóteses — e se manifesta com base em probabilidades. Como no exemplo a seguir:

- Mulheres portadoras de uma mutação no gene X provavelmente desenvolverão câncer.
- Maria é uma mulher.
- Maria é portadora de uma mutação no gene X.
- Maria provavelmente desenvolverá câncer.

A estrutura se assemelha à do raciocínio dedutivo. No entanto, diferentemente da dedução, cuja conclusão é necessariamente verdadeira quando as premissas são verdadeiras e o argumento é válido, o raciocínio indutivo estabelece a hipótese com base em uma série de regularidades (a maioria das mulheres desenvolve câncer, pois isso ocorreu diversas vezes no passado) e chega a uma conclusão probabilística. Além disso, diferentemente do raciocínio dedutivo, a afirmação “mulheres portadoras de uma mutação no gene X provavelmente desenvolverão câncer” não constitui uma lei da natureza que estabeleça uma regra geral e verdadeira (como “a mutação no gene X sempre provocará câncer”), mas expressa uma probabilidade.

Outro exemplo de raciocínio indutivo, que admite falhas: no caso de um homem que possui um alelo para olhos castanhos e outro para olhos azuis, a probabilidade de seu filho ter olhos castanhos é, teoricamente, de 50%. Essa probabilidade pode não se concretizar dessa forma caso existam fatores inerentes à dinâmica genética que favoreçam uma característica ou outra — e a ciência pode descobrir isso ao aprofundar suas investigações.

Depois de ler os autores clássicos de Filosofia da Ciência, fica muito forte a impressão de que há muito pouco a acrescentar a esse campo do conhecimento. O último livro que li sobre o assunto foi “Patterns of Discovery”, de Norwood Russell Hanson (minha resenha aqui: https://www.goodreads.com/review/show...), que me parece ser a raspa do tacho, dada a importância que atribui à observação dos dados e às atividades realizadas antes da formulação de uma hipótese, bem como a ênfase que coloca na inferência lógica abdutiva ou retrodutiva de Charles Sanders Peirce.

Hempel não menciona a retrodução, pois se concentra nas figuras mais conhecidas da dedução e da indução. Sua explicação dessas duas formas de inferência está, contudo, entre as mais claras.

Assim como o livro de Hanson, “The Philosophy of Natural Science” de Hempel carece de uma conclusão ou, pelo menos, de um resumo dos temas abordados, que certamente melhoraria toda a experiência de leitura.

[Allan Olley · Rating 5 out of 5]

This short volume is an introduction to the philosophy of natural science. It is part of a series the "Prentice-Hall Foundations of Philosophy Series". It offers an interesting survey of the then basic concepts of the field and some of the controversies around them. Much of the text is taken up with how scientific hypotheses are formulated and tested, with examples from history including Semmelweis's development of antiseptic practices in medicine and Pascal's experiments with mercury barometers to establish theories of atmospheric pressure. There is also extensive discussion of ideas of scientific explanation and in particular reduction. The discussions while brief and often in a very different context from later debates in the field still often nicely introduce and talk around various positions in the field such as realism and antirealism about the contents of theories. Hempel's own ideas about explanation and reduction are succinctly presented here.

The book contains an interesting further reading section. Also the in text references can be interesting, and includes a very brief mention of Kuhn's Structure of Scientific Revolutions that would significantly reshape the field.

[Folkert Wierda · Rating 3 out of 5]

Again a study book I read 35 years ago. Hempel is old-school standard model philosophy of science. But I must admit that he shows a decent level of relativism, he is not preaching nor black and white in his prescriptions regarding the scientific process. He discusses hypothesis forming, concepts, theory and briefly, really briefly, in the last chapter, the discussion about reductionism.
Hempel doesn't have a very colorful style of writing, and he presents the topic in a quite abstract fashion. I assume there are better introductory texts if you need to get into it.

Folkert
Www.ExcitingTechnology.net

[Ryan · Rating 5 out of 5]

Philosophy of Natural Science was a delightful balance of brevity and depth. It was a lucid as well as rigorous account of the foundations of scientific thought and method. I found wonderful insight into the nature of scientific hypotheses and theories.

[Reader2007 · Rating 4 out of 5]

Science, Faith, and Technology.

Really confusing, and could be much less wordy. Conclusions are good, though!

[Demian · Rating 4 out of 5]

Hempel te como la boquita. Primer libro de filosofía de la ciencia que me leo. Super interesante (y accesible)

[Kassidy Wheeler · Rating 4 out of 5]

I enjoyed this book more than Laudan, probably because I read it earlier in the semester (lol). The way he broke down how Semmelweis examines the infant mortality rate of a hospital in one division compared to another and uses logic to conclude that Doctors need to wash their hands was the most interesting part of the book to me. It is a challenging read if you are not deeply into philosophy and science.

I'm not sure I agree with some notions I derived from the book. For instance, Hempel argues that if Testability cannot predict or explain, then it is not scientifically useful. I do not agree with this because we can at times look at our failures and shortcomings we can still learn procedurally from these outcomes or we can learn how to derive the correct answers. Hempel relies solely on explanatory relevance and testability, whereas I suppose ad hoc hypotheses are lazy, they still could lead to interesting results that might explain some other phenomenon, I don't know. Hempel argues that ad hoc hypotheses are generated for the sole purpose of saving an evidentially threatened hypothesis, not called for by other findings, which then leads to no additional test implications. I suppose he's correct but is this not a more traditional way of philosophical thinking? Maybe we shouldn't save hypotheses if they are incorrect or invalid?

[Keyla · Rating 4 out of 5]

Filosofía de la ciencia natural de Carl G. Hempel es una introducción clara y rigurosa a los principales problemas de la filosofía de la ciencia. Representante del empirismo lógico, Hempel expone con precisión temas como la naturaleza de las hipótesis científicas, la función de las leyes generales, el papel de la observación y la importancia del lenguaje lógico en la construcción del conocimiento. Su célebre modelo nomológico-deductivo, que entiende la explicación científica como la deducción de un fenómeno a partir de leyes y condiciones iniciales, se presenta aquí de manera accesible para estudiantes y especialistas.

Aunque la obra refleja los límites del positivismo lógico, sigue siendo fundamental para comprender los debates contemporáneos, pues permite apreciar cómo se pensaba la ciencia en el siglo XX y qué críticas posteriores abrieron nuevos caminos. En suma, un texto breve, sistemático y altamente recomendable, que merece una sólida calificación de cuatro estrellas

[Thomas Maher · Rating 4 out of 5]

Very clear and well-organized, super useful pedagogically. Good for assigning intro readings on induction, confirmation, and the deductive-nomological model of explanation. Students seem to find the Semmelweis case study in Ch. 2 very helpful in understanding naive inductivism and its problems, as well as the context of discovery/context of justification distinction. Discussions of operationalism in Ch. 7 are a nice springboard for discussions of verificationism/empiricism. Thinking of also maybe throwing in Ch. 8 and using it as an introduction to Nagelian reduction, if I ever teach HPS again.

[tea · Rating 4 out of 5]

ja sam ovo čitala zbog ispita iz filozofije nauke ali moram priznati da mi nije žao što sam pročitala! pošto je meni ovo bio prvi put da se susrećem sa filozofijom nauke, rekla bih da je ovo skroz ok za stvaranje slike neke početne! ipak, 4 zvezdice jer bilo puno pozivanja na neke koncepte iz prirodnih nauka (baš čudno, pa knjiga se zove filozofija prirodnih nauka!) ali na način kao da se pretpostavlja da se to već zna.. a ja eto mnogo toga od ponuđenog nisam znala pa onda guglaj i slično. zato 4⭐ inače, kao osnova super a i stil je ok, za razliku od kunovog npr.

[Juan Bascur · Rating 3 out of 5]

La palabra "filosofía" es empleada aquí como "un conjunto sistemático de razonamientos".

En otras palabras, es un tratado del método científico y sus formas, con ejemplos excelentes e interesantes para mantener la atención del lector cuando el tema se pone demasiado denso.

A veces, se siente que Hempel se ensaña demasiado con otros autores científicos. Si se lo merecen o no, queda fuera de mi jurisdicción.

[Alberto Vidal · Rating 5 out of 5]

Hempel es uno de los filósofos de la ciencia más relevantes del siglo XX, y uno de los iniciadores a su vez del empirismo lógico, corriente filosofía madura de la llamada Concepción Heredada de la Ciencia.
Este obra tiene un carácter puramente divulgativo, tratando de forma didáctica la concepción que propone de la ciencia. Hempel defenderá el método hipotético-deductivo (nomológico deductivo) como método propio de la ciencia, y expondrá una serie de cuestiones relacionadas especialmente con el tema de la contrastación y la explicación.

[Dennis Robbins · Rating 5 out of 5]

An important and influential work in the philosophy of science. Hempel described the empirical sciences as containing the natural sciences (physics, chemistry, biology) and the social sciences (psychology, sociology, economics). At the end of his book he describes the reducibility of the social sciences to physical-chemical concepts. Most of the book, however, is devoted fundamental aspects of science such as experimental methods, hypothesis testing, requirements of explanations, general characteristics of theories and much more in a short book.

[emma]

if i had to wade through this whole book i am damn well counting it towards my reading challenge

[Krysztyna · Rating 2 out of 5]

Bardzo przegadana i bardzo fizyczna

[Lulu]

Science

[C · Rating 4 out of 5]

Aspects of Scientific Explanation is better (because more in-depth & developed analysis of same issues). This might be a good intro book though.

[Juan]

Va, pero entonces la astrología cuenta como ciencia o no?

[Michael Maxey · Rating 3 out of 5]

Useful, but not a life changing text. A resource I may come back to, but one that is just acting as a consolidation piece and partially seeing things from a different point of view.

[David Norcross · Rating 2 out of 5]

Insightful but Kuhn btfos

[Linda · Rating 3 out of 5]

Lihtne raamat mida kirjutati liiga keeruliste sõnadega...

[Leila · Rating 3 out of 5]

Hempel's analyses zijn nauwgezet en systematisch, en bij tijden erg verhelderend, maar hij maakt er te veel woorden aan vuil.

[Naitirofanboy · Rating 3 out of 5]

Capítulo 5. LAS LEYES Y SU PAPEL EN LA EXPLICACIÓN CIENTÍFICA.

[Lysergius · Rating 3 out of 5]

Exhaustive