How can we redirect ideas from the philosophy of science (which has focused in the past on physics, chemistry, astronomy, biology, etc…) to interrogating computer science or what Herbert A. Simon called The Sciences of the Artificial (as opposed to sciences which study natural phenomena)? Many scholars take the view that computer science – despite its name – is not even a science, it is a technology. Terms like science and technology are somewhat obsolete. They belong to modernism, and we are now living in hyper-modernism. We need to invent newer terms. Starting from what Paul Feyerabend wrote about the history and philosophy of science a few decades ago, I develop the thesis that a certain portion or kernel of computer science is scientific (a term that I would like to slowly displace towards a newer term) and much of computer science is cultural (also a placeholder term to be refined later) and changes in paradigm from decade to decade.
Alan Turing’s and John von Neumann’s original World War II-era computer science is not the same as 1960s COBOL business-procedural programming nor the same as 1980s object-orientation and the Xerox PARC (where Steve Jobs and Bill Gates pilfered their ideas) graphical user interface revolution, and then Artificial Life, quantum computing, biological computing, neural networks, Deep Learning, etc. These are all different paradigms of computing. And then the new paradigm Creative Coding and whatever we wish to make of it. Anything goes! Understanding the history of computer science and its changing paradigms as divided between scientific and cultural components (and my claim is that this two-level configuration is already evident within the work of Turing and von Neumann) is necessary to lay the groundwork for freeing the future of informatics or digital technology or Creative Coding as an existentially open-ended project of art and culture, or ethics and expressivity, where we can both respect science and change the world.
The radical University of California, Berkeley philosopher of science Paul Feyerabend (whose slogan was “Anything goes!”) receives much less attention nowadays than other major figures in science and technology studies such as Bruno Latour, Donna Haraway, and Karen Barad. Feyerabend receives less attention in the more specifically defined field of the philosophy and history of science than, say, Thomas S. Kuhn or Simon Schaffer. Feyerabend is important in a chapter of the history of ideas which was the paradigm shift in how we think about science that took place in the 1960s-1970s. He is sort of the anti-Karl Popper. To simplify, Popper was an extreme positivist who believed that scientific rationalism is a zone of high rarefied air that is above all possible contemplations of the biographical, cultural, economic, institutional, and historical contexts in which any given scientific discovery occurs. Popper dominated the field of the philosophy of science during the first half of the twentieth century. Then in the 1960s his worldview and influence crashed. Feyerabend was Popper’s academic assistant, and then rebelled against him, and rightly so. Seen in the context of this rebellion – and perhaps Oedipal struggle – there is something pure and lucid about Feyerabend’s vision.
The defenders of science in the science wars which began in the 1990s have attacked humanities disciplines such as science and technology studies for their supposed relativism. In the self-image regarding their work of the natural scientists themselves – and in the mid-twentieth century philosophy of science of Karl Popper – science is allegedly rooted in rationality and logic and is the pursuit of time-independent or history-independent truths. I articulate a third position between the two poles of rational-and-objective versus culturally influenced. I believe that both dimensions are valid – scientific and cultural. I argue that computer science or informatics (as an example of a techno-science) has one dimension which is rational-objective and time-independent and one dimension or layer which is time-dependent.
What is objective and eternal as science in Alan Turing’s 1936 formulation (and related formulations during the birth of computer science which soon followed) is the encoding and physical writing on temporary memory or a storage medium of both programs and code as binary numbers. What changes with time – from decade to decade in the history of software coding paradigms and can be called cultural – is the relationship between code and data. The relationship between code and data has many different historical configurations. This relation changes in technical-technological paradigm shifts in parallel with shifts in broader socio-cultural paradigms (which influence the era-specific purposes for which computers are utilized). Early computers were deployed for scientific calculations, and for manipulating numbers using logical rules. Alan Turing’s milestone achievement was partly scientific and partly cultural.
Positivism in the Philosophy of Science
During the first half of the twentieth century, and more specifically in the period extending from 1921 (the year of the publication of philosopher of language Ludwig Wittgenstein’s Tractatus Logico-Philosophicus) until about the 1960s, the understanding of the relationship between science and philosophy (or the ascending field of the philosophy of science) at universities in the Western countries was dominated by the school of thought known as logical positivism or logical empiricism, also known more broadly as British analytical philosophy. The movement of positivism experienced an invigoration and formalization with the appearance of the Vienna Circle, a group of scientists, mathematicians, and philosophers who convened regularly from 1924 to 1936 at meetings at the University of Vienna chaired by the logician Moritz Schlick.
The logical positivists proclaimed science and the scientific method of empirical observation, experimentation, formulation and testing of hypotheses, skepticism, and reasoning to be a superior form of knowledge to the humanities, the arts, and philosophy – superior to any so-called metaphysical speculation about (for example) the meaning of life, the place of humanity in the universe, the debate over free will and determinism, or the existence of God. Non-scientific philosophical works were dismissed as essentially meaningless. They were regarded as mere flourishes of language dealing with questions which have no answers. The mode of cognition known as science, according to the hegemonic logical positivist thinkers of the Anglo-American academy, renders philosophy obsolete. Some of the principal early twentieth-century thinkers who catalyzed the movement were Gottlob Frege, Bertrand Russell, and Rudoph Carnap. A.J. Ayer’s Language, Truth and Logic (1936) was a milestone work in England that provided an overview of the movement’s Weltanschauung.
Karl Popper’s Ambivalent Orthodoxy
The twentieth-century orthodoxy in the philosophy of science was established, and maintained for a long time, by the work of the Austrian-British scholar Karl Popper. Popper expounded his position most extensively in his 1959 book The Logic of Scientific Discovery, which had been published earlier in a shorter version in German in 1934 as Logik der Forschung. Zur Erkenntnistheorie der modernen Naturwissenschaft. Contrasted with the subsequent 1960s and beyond generation of philosophers of science who are represented in the present study by Thomas S. Kuhn, Paul Feyerabend, and Simon Schaffer, Popper still privileges the rational and objective logic of science over any considerations of the historical, cultural, political-economic and institutional contexts in which science develops.
It should be noted, however, that Popper is generally seen as ambivalently being a critic of Vienna School logical positivism. The great achievement of his book is his critique of the centuries-old inductivist view of the scientific method, and his advocacy of the alternative approach of empirical falsification. Inductivism was the rendition of scientific method which the philosopher Francis Bacon had formulated in the seventeenth century. It underlines the growing acceptance of a scientific truth as the process advances of further observations strengthening the case for elevating a modest law into a broader law. This progression contributes to the increasing general comprehension by scientists of the true causal structure of nature. Karl Popper asserted, to the contrary, that theories can never be completely proven, but they can be disproven or refuted. Popper’s innovation of a cardinal principle of falsifiability marks indeed a step towards the deconstruction of the traditional orthodox version of science and its eventual supersession by the hyper-modernist position outlined in the present study. His emphasis on the disproving of scientific laws moves in the direction of Kuhn’s idea of the paradigm shift in the latter’s The Structure of Scientific Revolutions (1962).
In his 1975 essay “The Rationality of Scientific Revolutions” (based on a lecture delivered at Oxford University in 1973), Karl Popper examines progress in science from an evolutionary or biological point of view, advancing what he calls genetic and behavioral metaphors. He goes on to address the question of progress in science deploying strictly rational or logical criteria. He concludes with a section on extra-scientific obstacles to scientific progress. Popper puts forth the idea that any new valid scientific theory must overthrow or controvert its predecessor theory (or guiding explanation of the object of inquiry) within the domain at hand. A crucial example to consider is a monumental episode in the history of theoretical physics: Popper notes that Einstein’s theory refutes Newton’s theory. Scientific progress is, for Popper, always revolutionary. Popper also asserts – and he is adamant about how decisive this is for his perspective – that Newton’s and Einstein’s respective theories are not incommensurable. The parallel claims of overthrowing and commensurability would seem to be in contradiction with each other, but they are not. Yet Popper fails to explain the reconciliation of the two apparently contradictory ideas subtly enough.
One could say that Newton’s theory is still valid and useful today. It is still taught to high school students as the basics of physics. Or, as Kuhn says: “Relativistic dynamics cannot have shown Newtonian dynamics to be wrong, for Newtonian dynamics is still used with great success by most engineers and, in selected applications, by many physicists.” Newtonian physics is an approximation of the true nature of reality that is efficacious within the confines of a certain perceptual framework: the familiar circumstances of everyday life where the extreme dimensional phenomena of the vastness of outer space and the tininess of subatomic particles, or the relativistic spacetimes associated with extreme speeds, do not need to be considered. One would need to exit the exalted sacrosanct world of science proper – to the social and historical backgrounds and contexts of science – in order to make intelligible the ambivalent character of both truth value and distance from “the truth” of the Newtonian approximation. Popper refuses to venture outside of the lofty rarefied space of what has been anointed as objective knowledge. This refusal is the whole point of what Popper is on about, the dam that is holding back the invading waters of socio-cultural transdisciplinary investigations of the pure rational-logical cognitive territory claimed for science.
The Open Society and Its Enemies
Progress in science, according to Popper, can be assessed utterly rationally, as it is accessible to evaluation by impartial tests and criteria. He writes: “It is rationally decidable whether or not a new theory is better than its predecessor.” For Popper, scientific revolutions have nothing to do with power relations or political economy. They have nothing to do with what Karl Marx believed revolutions to be. The personal biography of the scientist is irrelevant for Popper. The pioneering scientist may have been inspired by an epiphany or by dramatic psycho-biographical events which engendered his genius, but this means nothing for Popper’s interpretive scheme, since only the results of research and thought experiments are what count.
In the final part of his lecture on the rationality of scientific revolutions, Popper talks about sociological, economic, and what he calls ideological obstacles to progress in science. Popper was also a preeminent twentieth-century liberal social theorist, the author of the acclaimed work The Open Society and Its Enemies (1945). Yet his work in the philosophy of science and his work in political philosophy remain entirely separated from each other in a strict (Kantian) dualism (similar to the unfortunate dualism of science and politics in the system of thinking of Bertrand Russell and of his decades-later American “disciple” Noam Chomsky).
Popper’s interest in the sociology of science is restricted to reflections on those extra-scientific factors – such as intolerance or dogmatism – which inhibit the progress of science. What can also do damage to science is precisely the opposite – when a scientific theory captures the imagination of the public so grippingly that the theory is threatened or ruined by an excess of success. The theory becomes an intellectual fashion. Popper calls this the degeneration of a scientific revolution into an ideological revolution. He cites the examples of the fate in popular and media discourse of the Copernican and Darwinian and Einsteinian and quantum physics scientific-becoming-ideological revolutions. The limitation of Popper’s framework is that he is only able to contemplate the sociological dimension as an obstacle.
The open society as concept was first suggested in 1932 by the French philosopher Henri Bergson. The concept of open is a promising beginning towards developing a legitimate successor concept to public. We must not understand open in the simplistic sense of giving everything away for free (freeware), but instead understand it in the sense elaborated by the liberal political philosopher Karl Popper in his 1945 book The Open Society and Its Enemies. Popper conceptualized the open society as a short-lived transitional phase between the organic, tribal, or closed society and the abstract, depersonalized, bureaucratic society where much is decided by rules, regulations, and automatic processes. In that nightmare dystopian society, there is a lack of face-to-face transactions and engagements among individuals. The open society, by contrast, is characterized by a critical attitude towards tradition, and by a collective awareness that individuals make moral decisions. The open society is anti-authoritarian. As the English-language Wikipedia article on Karl Popper elaborates, the government of the open society is responsive and tolerant, and its structures are transparent and flexible.
In the philosophy of science, starting in the 1960s, there appeared important scholars who placed into question the orthodoxy of Karl Popper. Three of these major figures are: Thomas S. Kuhn, Paul Feyerabend, and Simon Schaffer (and his sometimes co-author Steven Shapin). Kuhn emphasizes the concepts of scientific revolutions, normal science, paradigms, and paradigm shifts. Feyerabend argues against claims to universality of how the scientific method is or should be practiced, pleads against pure facts and for an observation language, and elaborates his anarchist epistemology. Schaffer and Shapin (whose work came much later than the 1960s) underline the historical contexts of experimental scientific research and the legitimacy at the time (the moment when it appeared to be valid) of knowledge methods in episodes of scientific history which later come to be judged as pre-scientific and antiquated.
The works of these three luminaries have been controversial and have been much debated over the course of time. Kuhn was attacked for alleged relativism (rejection of the rationality and objectivity of science). Feyerabend was called “the worst enemy of science.” Schaffer and Shapin were criticized for their supposed limitation of being mere historians who view scientific discourses as strategies of power. The ideas of Kuhn, Feyerabend, and Schaffer question the fundamentals of our image of what science is, and these authors sparked much contentious debate in the past. The fact is that many commentators and readers have liked their radical ideas and many others have disliked them. Rehashing their ideas is in itself unlikely to change anyone’s mind. I want to be clear about what is the point of my revisiting their ideas.
The earlier debate of a couple of decades ago came down to this question: do these radical views – of Kuhn, Feyerabend, and Schaffer and Shapin – describe accurately how science works and proceeds? This famous debate is essentially over and obsolete by now. If I had to take sides, I would certainly be on the side of defending the ideas of those three thinkers as accurate in their portrayals of pre-modern and modern scientific discoveries – such as the revolutions in astronomy, physics, and mathematics of Ptolemy and Copernicus and Galileo; or Robert Boyle’s seventeenth century air pump; or Einstein’s special and general relativity. But my defense of these philosophers of science in this way would not convince anyone. I instead undertake something quite different. I develop an original position in my remarks on Kuhn, Feyerabend, and Schaffer and Shapin. Instead of maintaining that their ideas describe science as it has been in the past with veracity, I argue that science has changed and that it has now caught up with their prophetic views. They were pioneers and ahead of their time.
Thomas Kuhn, The Structure of Scientific Revolutions
Kuhn begins his seminal 1962 work The Structure of Scientific Revolutions by acknowledging the crucial influences on him of two thinkers in the history and philosophy of science – Alexandre Koyré (Études galiléennes of 1939 and From the Closed World to the Infinite Universe of 1957) and Ludwig Fleck (Genesis and Development of a Scientific Fact of 1935). Kuhn tells us in the Preface to his book that his main concern is going to be a focus on the historical and experimental conditions for the emergence of a new theory or discovery in science. He explains that the critical experience in his biography which was the genesis of his theory of alternating scientific revolutions and normal science practices was his intense perception that much livelier discussions occur among scholars in the social sciences than in the natural sciences. Practitioners in the natural sciences who go about the business of what Kuhn calls puzzle-solving generally do not think about or discuss fundamentals of their field. One might say that someone else already did their thinking for them.
Kuhn’s historical outlook is the opposite of the epistemological orientation of Popper, who believed that only the end results, the scientific achievements which become part of the rational-logical scientific canon, count as important. Kuhn wants science to include the supposedly out-of-date theories which have been superseded. He defends the “historical integrity of that [discarded] science in its own time” and raises the question of commensurability or incommensurability between different scientific theories, an area of inquiry which has become a notable topic in the philosophy of science.
In the Introduction to The Structure of Scientific Revolutions, Kuhn asserts that the activities of scientists during the periods of normality are “predicated on the assumption that the scientific community knows what the world is like.” What goes to a large degree uninterrogated in Kuhn’s work is a philosophical reflection on the concept of world. What exactly is the world? The rational and moral mission of the natural sciences is usually declared as being that of observing, researching, and understanding the true and objective nature of the world (or of so-called reality). The object of inquiry of physics is the physical world (or the universe). The object of inquiry of biology is the world of life and living organisms. Chemistry studies the elements (of the periodic table) and the compounds made up of atoms and molecules which comprise the matter, substances, and energy of the world. Earth science studies the properties and processes of the world as planet: Earth and its atmosphere, ecosphere, oceans, geologies and geographies, and its crust and upper mantle. Astronomy has its origins in the ocular and then visual-media-supported telescopic observations of the night sky, a nocturnal experience which is arguably part of the framework of our perceptual relationship to the world.
What is Normal Science?
Computer science is thus ironically the continuation of science. It is right at home with what science always was – yet this was obscured by our mythology of science. To return to Thomas S. Kuhn and The Structure of Scientific Revolutions: Beyond what Kuhn famously says about normal science, paradigms, paradigm shifts, and scientific revolutions, he is also profoundly saying something else. Kuhn makes clear philosophically that there is no immediate experience of the world – there are only frameworks of interpretation. The successive scientific paradigms are frameworks of perceptual knowing and elucidation. Kuhn writes: “Philosophers of science have repeatedly demonstrated that more than one theoretical construction can always be placed upon a given collection of data.”
Kuhn goes on to explain in The Structure of Scientific Revolutions that normal science is the research activity based upon past scientific achievements, and that paradigm and normal science are two closely and intricately related terms. The adherents to a previous paradigm convert step-to-step to the new paradigm. The older school gradually disappears. Kuhn writes: “Paradigms gain their status because they are more successful than their competitors in solving a few problems that the group of practitioners has come to recognize as acute.” The work of normal science is essentially what he calls mopping-up operations, carrying out tasks to bring closer together various facts and details with the predictions made by the paradigm.
Normal science concentrates on three areas. The first area is the class of facts that the new paradigm has shown to be especially revealing of the nature of things, as it is henceforth to be understood. The new paradigm has made these facts worth identifying both with more precision and in a larger variety of situations. A second area or class of facts are those which can be brought in line directly with predictions flowing from the new scientific theory. This is called the demonstration of agreement. Kuhn writes: “Often the paradigm theory is implicated directly in the design of apparatus able to solve the problem.” A third class of experiments and observations is empirical work undertaken to further articulate the paradigm theory. This class turns out to be the most important of all. It is necessary in order to arrive at a clear definition of the problem and to guarantee the long-term existence and stability of the solution.
Normal science engages primarily in puzzle-solving. Its practitioners do not aim to produce any major novelties, neither conceptual nor phenomenal (or they might seek out breakthroughs and rarely succeed). The fascination of normal science research lies not in achieving what has already been anticipated, but in discovering new pathways towards the already known goal and its repeated verification. It requires the solution of all sorts of complex instrumental, conceptual, and mathematical puzzles. He who succeeds in this area of work proves himself to be an expert puzzle-solver. The community of scientists operating within this paradigm encourages its members only to tackle challenges which have a known solution in advance. Other problems are rejected as metaphysical or as the proper domain of another discipline. Only those scientific questions are to be pursued which can be formulated with the concepts and addressed by the instruments available within the paradigm. There exist rules which limit the scope of acceptable solutions and by which one is allowed to arrive at those results. Kuhn’s assessment of normal science remains ambivalent: professionalization leads to rigidity and restricts the scientist’s vision, yet it inspires an intense emphasis on acquiring detailed information and accomplishing observational precision.
Here Comes the Paradigm Shift
Kuhn asserts that, in the history of science, discoveries (novelties of fact) and inventions (novelties of theory) are not so distinct from each other. Important scientific discoveries that incite paradigm shifts belong generally to an era of history and cannot reasonably be attributed only to a specific individual scientist or a single date in time. A new paradigm in any given science does not necessarily render the previous paradigm invalid. Copernican astronomy appears to have superseded the astronomical system of Ptolemy – yet the calculations and predictions of the ancient Greek-Egyptian mathematician were robust and are still widely used today in engineering contexts. The heliocentric discoveries of Copernicus in the sixteenth century and Galileo in the seventeenth century ignited a kind of metaphoric supernova explosion. The Copernican model of the sun-earth relationship, which disputed and eventually supplanted the geocentric universe of Ptolemy, was not accepted for centuries due to the anxiety about the loss of our anthropocentric status in the cosmos which it provoked. Humans, created in God’s image, were no longer the center of the universe. The sun does not revolve around the earth as was previously believed; the earth revolves around the sun. Physical reality and its classical laws were elevated to a sovereign status in relation to humans.
Kuhn describes how the beginning murmurs of a paradigm shift start to become audible. Anomalies or counter-instances to the prevailing theory occur in the crisis phase. The decision to reject the prevalent paradigm is simultaneous with the decision to embrace the new one. After an interlude of resistance, encompassing various attempts to resolve the crisis quickly through modifications to the existing theory, a change in framework or Gestalt metamorphosis of perception finds wide acceptance as the ultimate way to make sense of the new data. The crisis of an established scientific paradigm can end (in one possible scenario) through the normal science of that paradigm reasserting itself and maintaining its hold on the scientific community at hand; or the crisis coming to be seen as unsolvable (a second possible scenario) and no further resolution is sought in the short term; or finally (in a third scenario), a new candidate for paradigmatic dominance emerges and a battle for hegemony ensues. During the transition period there is an overlap between the approaches to problems of the old and new paradigms.
In his 1969 Postscript to The Structure of Scientific Revolutions, Thomas S. Kuhn points out that he intended two different meanings in the book for the term paradigm. The first meaning is the entire constellation of group commitments (ideas, tools, and research methods), values, beliefs, and techniques shared by the members of a given scientific community. The second meaning refers to only one element of that constellation: the concrete solutions to puzzles that are encountered in practice, and which end up being shared models or examples of how to apply the consensus theories according to an agreed upon set of rules.
Feyerabend’s Reversal of Theory and Observation
Paul Feyerabend argues that scientific theories are, and, in fact, ought to be, inconsistent with one another. These incompatibilities accrue due to the very nature of terminological systems of explanation and description. He asserts that “the meaning of every term we use depends upon the theoretical context in which it occurs. Words do not mean something in isolation: they obtain their meanings by being part of a theoretical system.” Science is, according to Feyerabend, the continuous search for an appropriate observation language: an idiom which forges as directly as possible a fruitful relationship to facts and data.
The Austrian philosopher who taught at Berkeley often says that things in the history and methodology of science are really the opposite of what they are usually claimed to be. His position implies a reversal in the assumed relationship between theory and observation. The significations of theoretical terms do not derive from an a priori language of an essential naming of the world, as this was claimed by the logical positivist tradition of the Vienna Circle and Karl Popper. The observation terms, like any other phraseology, depend for their meaning on the theoretical context in which they occur. The observational statement, for Feyerabend, is in need of interpretation and does not have a self-evident meaning, as the traditional empiricist philosophers of science had alleged. Theories can be meaningful on their own without observations; an observation devoid of a theory, however, cannot attain to stability or sense.
From these insights, Paul Feyerabend concludes that “an adequate empiricism itself therefore requires the detailed development of as many different alternative theories as possible, and this is the methodological justification of a plurality of theories. A scientist or philosopher must be allowed to start completely from scratch and to redefine completely his domain of investigation.” There is no theory-independent observation language. The theory should emerge slowly and immanently from the experience.