Mount St Vincent University
In many recent (and some not so recent) publications Noam Chomsky makes an appeal to Galilean science and claims the Galilean framework justifies his own approach to scientific inquiry (e.g., Chomsky, 2002, 2009, 2010, 2012). Allegedly, this approach has a distinguished scientific and philosophical tradition. “Chomsky’s science of language is a science in the Cartesian-Galilean tradition. It is a branch of the study of biology” (McGilvray 2005: 4). In this blog post I argue that this approach should be rejected because it rests on a superficial and incorrect interpretation of Galileo’s work, has been rejected already by Rene Descartes, and is contrary to established scientific practice.
Without a doubt, Noam Chomsky is the best known linguist and his success has been linked to his persuasive debating style and his emphasis on rigorous scientific methodology for linguistic research. Yet, over the years Chomsky’s attitude towards the scientific method has changed, and he acts now as if no data can challenge his own proposals. For example when asked what kind of empirical discovery would lead to the rejection of the strong minimalist thesis, Chomsky replied: “All the phenomena of language appear to refute it” (Chomsky, 2002, 124, emphasis added). Yet, he is not willing to abandon the minimalist thesis. Instead he suggests dismissing the data that seem to challenge it. Chomsky claims that such a large-scale dismissal of data that are inconvenient for his view is based on a “Galilean style… [which] is the recognition that…the array of phenomena is some distortion of the truth … [and] it often makes good sense to disregard phenomena and search for principles” (Chomsky, 2002, 99). Chomsky calls this attitude the “Galilean move towards discarding recalcitrant phenomena” (Chomsky, 2002, 102). He claims that massive data dismissal was advocated by Galileo: “[Galileo] dismissed a lot of data; he was willing to say: ‘Look, if the data refute the theory, the data are probably wrong.’ And the data that he threw out were not minor” (Chomsky, 2002, 98). He then proposes that is was accepted by other famous scientists (e.g., Copernicus, Newton, Einstein, Monod) and that it “is pretty much the way science often seems to work …You just see that some ideas simply look right, and then you sort of put aside the data that refute them” (Chomsky, 2009, 36). Data-dismissal has been advocated numerous times in Chomsky’s publications, culminating in the argument from the Norman Conquest: “… if you want to study distinctive properties of language – what really makes it different from the digestive system … you’re going to abstract away from the Norman Conquest. But that means abstracting away from the whole mass of data that interests the linguist who wants to work on a particular language” (Chomsky, 2012, 84, emphasis added). Arguably the most bizarre invocation of the Galilean style occurs when Chomsky suggests: “…if we want a productive theory-constructive [effort], we’re going to have to relax our stringent criteria and accept things that we know don’t make any sense, and hope that some day somebody will make some sense out of them” (Chomsky, 2012, 169).
Chomsky is not the only defender of the Galilean style. It has been suggested that “[a] significant feature of the Generative Revolution in linguistics has been the development of a Galilean style in that field” (Freidin & Vergnaud, 2001, 647). Attaching the label Galilean to a style of inquiry suggests two things. First, it implies that Galileo worked using the same or a very similar style. Second, given the massive success of the Galilean scientific revolution, it suggests that work adopting a Galilean style is superior to (all) other work. While the second suggestion seems uncontroversial, the first needs support from the actual work of Galileo. Below I argue that the success of Galileo as a scientist was not based on a massive dismal of data that were inconvenient to his theories and that he would have rejected proposals that required him “to accept things that make no sense”.
1. The Feyerabendian/Chomskyan interpretation of the Galilean Style
Interpretation of the Galilean method is notoriously controversial. One historian of science remarks: “Hardly any other icon of modern science has become as much a victim of his interpreters as Galileo” (Fischer, 1992, 165). Many interpretations were not based on impartial evaluation of the original work but on fashionable historicism or on the desire to bolster one’s own methodology by appeal to Galilean authority. Interpreters attribute to Galileo views from inductionism (Mach) to rationalism (Cassirer, Lewin) and methodological anarchism (Feyerabend).
Feyerabend alleges that Galileo replaced a well-established philosophical/scientific system and violated rationally grounded methodology to justify a worldview (heliocentrism) that clashed with observable data. He attributes the success of the Galilean model to the invention of auxiliary hypotheses and fraudulent tricks that compensated for lack of explanatory power of the Copernican theory (the centerpiece of Galileo’s world view).
Dogmatic insistence on a falsified central hypothesis, the development of a scaffolding of auxiliary hypotheses from related sciences, and reinterpretation of problematic evidence allowed the emergence of a new system of natural science that replaced the old… Galileo’s success became possible because he proceeded pragmatically, without feeling bound by any rules of rationality or methodology.
(Fischer, 1992, 168)
Chomsky seems to share Feyerabend’s conviction that Galileo discarded results that did not fit his theories and that he insisted confidently on the correctness of his hypothesis. Possibly he even agrees with Feyerabend that Galileo employed fraudulent tricks to bolster his hypotheses. For Feyerabend one such trick is the use of the telescope. He claims that the use of this new instrument was problematic for several reasons. First, Galileo never ruled out that the images seen had been based on optical illusions. Second, it was not clear at the time that the instrument would work as reliably for astronomical observation as it did for terrestrial observation. Third, the rather primitive first telescopes produced images that were unfocused, contained distortions, coloured edges, halos, and other impurities. Therefore, it was difficult to determine which part of the image was caused by celestial bodies and which by imperfections of the telescope.
Feyerabend claimed that because of these problems Galileo could not base his theory on empirical observation but instead inferred the telescopic images from his theory. The picture that emerges is intended to suggest that Galileo violated not only the rules of the existing paradigm but all rules of rational inquiry. He is depicted not as brilliant scientist but as self-serving opportunistic propagandist: “Galileo was – simply said – the better charlatan and evangelist … He used rules in an opportunistic way to promote his ideas” (Fischer, 1992, 168). Feyerabend was deeply skeptical of the value of scientific inquiry and used Galileo’s success, based exclusively on methodological anarchism as one important case to support his anti-scientism. He claimed that, ultimately, science is just another form of religion and that theories attempting to establish the objective superiority of scientific methodology are a “previously not described form of insanity ” (Feyerabend, 1978, 7).
Chomsky seems to accept claims about the methodological anarchism of Galileo. But, unlike Feyerabend, he consider this anarchism a virtue, an essential component of rational scientific inquiry. He claims it is precisely the willingness to adjust the data to his hypothesis that allows him to make significant scientific progress and cite the far reaching success of the Galilean revolution as supporting evidence for this claim (e.g., Chomsky, 2002, 2009, 2012). This raises the question of whether Galileo’s success was indeed based on an anarchistic methodology. The next section answers this question.
2. Galileo, the (non-anarchistic) scientist
Chomsky seems to hold that the Galilean style is superior because it constituted a radical break from the accepted practices of the day and that this break allowed scientists to gain novel insights. The claim that Galileo’s science was a complete break from the then dominant Aristotelian science has been widely popularized by Thomas Kuhn. He holds Galileo’s new method exemplified a paradigm shift in science, making it incommensurable with the previous methodology: “Normal research, guided by [Aristotelian conceptual categories] could not have produced the laws Galileo discovered” (Kuhn, 1996, 123). Yet, it has also been suggested that the break was not as complete and dramatic but resulted from a gradual process of changes to scientific paradigms. Below I defend this second position.
Fischer suggests that Galileo was not overthrowing a well functioning system of natural philosophy but responding to developments beginning in the 14th century that had failed to address “endemic problems of the late scholastic syntheses” (Fischer, 1992, 168). The system of natural philosophy at Galileo’s time was an “eclectic and somewhat pluralistic conglomerate… proponents of different factions chose different approaches in response to new challenges” (Fischer, 1992, 169). In astronomy the new Copernican/Galilean system (developing a theory based exclusively on uniform circular motion) was in fact closer to Aristotelian ideals than the more complex, dominant Ptolemean system. Further, the Copernican theory was not in opposition to all existing approaches but had important predecessors in the work of some ancient thinkers (Aristarch, Seleucus, Philolaus, Plutarch), and some medieval scholastics (Buridan, Oresme). Presumably, there was also some influence from some Muslim Theologians (Fakhr al-Din al-Razi, Al-Zamakhshari, Ebussuud Efendi). Their proposals and hypotheses arose from within the Aristotelian framework, not in complete opposition to it. Thus, Copernicus and Galileo did not reject a single rational alternative to their own hypotheses.
Similarly, one can evaluate the validity of Galileo’s telescope-observations from a different perspective. The Galilean telescope may appear flawed and primitive from our perspective, but it was an important advance in technology in the 16th century. And it was continually improved, extending gradually the original (1609) 3-fold magnification (Drake, 1990, 133-4). Like any new technology there were certain risks in its use and uncertainties about its reliability. But the use of the telescope does not seem to be an instance of reckless methodological anarchism. Nor was Galileo the only one relying on the telescope. The new invention quickly gained popularity and was used by others who confirmed Galileo’s observations (Scheiner, Clavius). This is not to say that there were no problems with some of the observations and/or interpretations. Observations in 1610 led Galileo to the hypothesis that Saturn was accompanied by two small moons: “It is clear that Galileo assumes here wrongly that the blurred image is caused by a merging of the bright images of the members of a hypothesized three bodied system ” (Fischer, 1992, 170). Observations in 1612, when Saturn’s rings were edge-on from Earth’s perspective (seemingly disappearing), made him question this hypothesis: ‘I do not know what to say in a case so surprising, so unlooked for and so novel’. Further observations in 1616 and 1626 added evidence that challenged the three-body system hypothesis, and in 1655 Christian Huygens proposed after further careful observation that Saturn was surrounded by a solid ring. Thus, continued observation of the phenomenon resulted eventually in the correct hypothesis. Conflicting evidence was not ‘set aside’ but prompted further investigation which resulted eventually in the rejection of a hypothesis that had seemed reasonable based on the initially available evidence. It would be unreasonable to expect that every observation made leads straight away to the correct hypothesis. On the other hand, formulating hypotheses based on insufficient evidence can become a problem if scientists refuse to give up this hypothesis in light of new evidence.
Did Galileo use tricks and propaganda to support a hypothesis that had no solid theoretical grounding and conflicted with empirical evidence? In some cases this might have happened but in the context of Galileo’s entire body of work the defence of the Copernican world view is hardly an act of dogmatism in the face of empirical counter-evidence. Galileo’s physics had been developed prior to and independently of any Copernican commitments. They were based on existing knowledge and careful systematic experimentation. Galileo used Archimedean principles to solve problems in mechanics (hydraulics, movement of bodies, inertia etc.) and extended this approach gradually to new applications. At the same time he began his search for unifying mathematical principles. It is true that Galileo had great confidence in his method but this confidence was based on a systematic approach to well defined problems:
Where others speculate based on vague premises or arbitrarily changing suppositions Galileo makes deductions from evident and explicitly stated premises. Where others search for compromises Galileo strives for precise propositions and definite decisions. Where others appeal to authority Galileo demands evident axioms or conclusive empirical results.
(Fischer, 1992, 176)
Further, Galileo’s focus on mathematical analysis required conceptual precision. He understood that there is a difference between calculations based on ideal conditions and the observable movements of physical bodies. Galileo believed that this difference needs to be accounted for. However, it does not interfere with the ability to calculate based on general principles as long as it can be shown that it is accidental and insignificant. For Galileo it was important to separate accidental from essential properties. Finding essential properties allowed him to formulate hypotheses that make testable predictions. It took Galileo years of experimentation and theory refinement to arrive at proofs for some of the laws he had discovered.
While Galileo’s debating style might share self-confidence and assertiveness with Chomsky’s there are also considerable differences. When arguing with opponents Galileo first demonstrated that their assumption either led to logical contradictions or can be rejected on empirical grounds. Then, he introduced his own solutions and explained the mathematical and physical assumption that supported his premises. Next, he deduced consequences and showed how his conclusions could be supported empirically. Hypotheses that had been subjected to this testing procedure were assumed to be confirmed (demonstrated) but they were not irrefutable. It remained possible that further experiments provided refuting evidence. Many of the commendable aspects of the Galilean method are absent in the recent (post-2000) work of Chomsky. And, it is rather unlikely that Galileo would have advocated to ‘set aside’ recalcitrant evidence or to accept things that do not make any sense.
3. Descartes on the Galilean style
Given that Chomsky likes to call his method not only Galilean but also Cartesian (e.g., Chomsky 1966a, 2002, 2009, 2012), it seems relevant to consider Descartes’ opinion of the Galilean method. Descartes followed Galileo’s work with great interest. Especially Galileo’s trial and condemnation in 1633 had a profound impact on Descartes and he commented numerous times on Galileo’s work. However, Descartes’ critical evaluation of Galileo’s work has received only scant attention. Roger Ariew suggests this is the case because those who have commented on Descartes’ criticism “are generally in agreement that Descartes’ assertions lack value” (Ariew, 1986, 79). Ariew makes a convincing case rejecting the claims that Descartes arguments were defective, not meant for anyone but Mersenne’s attention, based on a desire to demean what did not conform to his own philosophy, or a reaction to unjustified plagiarism charges. Like Ariew I believe that Descartes gives an overall fair evaluation of Galileo’s work and raises some valid points of concern regarding the Galilean method. For example, Descartes deals with Galileo’s work in a letter to Mersenne (14/8/1634), providing his first reaction to The World: “I find that [Galileo] philosophizes well enough on motion, though there is very little he has to say about it that I find entirely true. As far as I could see, he goes wrong more often when following received opinion than when going beyond it” (CSMK III, 44). This comment indicates that Descartes values the original part of Galileo’s work as superior to that following the traditional method. Further, Descartes observes some flaws in the presentation of Galileo’s results: “The arguments which he uses to demonstrate the movement of the earth are very good; but it seems to me that he does not set them out in the way that is required if they are to be convincing, for the digressions which he introduces make you forget the earlier arguments when you are in the process of reading the later ones” (Ibid.). The concerns about methodological issues are more detailed in the following discussion (letter to Mersenne, 11/10/1638).
Generally speaking, I find [Galileo] philosophizes much more ably than is usual, in that, so far as he can, he abandons the errors of the Schools and tries to use mathematical methods in the investigation of physical questions. On that score, I am completely at one with him, for I hold that there is no other way to discover the truth. But he continually digresses, and he does not take time to explain matters fully. This, in my view, is a mistake: it shows that he has not investigated matters in an orderly way, and has merely sought explanations for some particular effects, without going into the primary causes in nature; hence his building lacks a foundation. Now the closer his style of philosophizing gets to the truth, the easier it is to recognize its faults, just as it is easier to tell when those who sometimes take the right road go astray than it is to point out aberrations in the case of those who never begin to follow it.
(CSMK III, 124)
This passage is important for two reasons. First, Descartes compliments Galileo on “philosophizing more ably than usual” and on using mathematical methods when investigating physical problems. This constitutes an improvement over the dominating scholastic approach. Second, the problems Descartes noticed earlier remain and lead to the accusation that Galileo’s building of reasoning lacks a foundation. In spite of some visible admiration for Galileo’s work, the criticism Descartes levels against his method is serious. Galileo has violated two “basic Cartesian methodological prescriptions ‘to consider a subject in an orderly fashion’ and ‘to explain a subject fully’” (Ariew, 1986, 83). That means Descartes is critical precisely of the aspects of Galileo’s work that Chomsky considers to be the foundations of the Galilean style.
Of course, Chomsky himself does not apply the Galilean style consistently. Many of his own ideas that “simply looked right” turned out to be wrong, including the following. In his early work Chomsky proposed that the base of a transformational grammar involves phrase structure (rewriting) rules (e.g. Chomsky 1957, 13-17; 1962, 127; 1975, 80); in later work these rules were abandoned (e.g. Chomsky 1986a, 82-83; 1986b, 3; 1995, 25). Chomsky first proposed that transformational rules are either optional or obligatory (Chomsky, 1957, 45; 1962, 136), then abandoned this proposal (Chomsky and Lasnik, 1977, 41), and claimed that transformational rules are all optional (Chomsky and Lasnik, 1977, 41). This proposal was subsequently abandoned (Chomsky 2000c, 130). In his early work, Chomsky proposed that a notion of well-formed (sentence) is fundamental to linguistics (Chomsky, 1957, 13-14; 1966b, 32; 1972, 64), but decades later he abandoned that proposal (Chomsky, 1995, 194, 213). In early work Chomsky insisted that the notion of deep structure is fundamental to NL syntax (Chomsky, 1966b, 91). Later he announced that language has no D-structure (Chomsky, 1995, 186-189). These are only a few of many instances in which Chomsky changed his mind about ideas that “just seemed right”. And he did this based on new data, which suggests that even in his own case the Galilean style has only very limited applicability. Finally, Chomsky does not accept the Galilean style for the work of others but rather strongly objects to massive dismissal of data: “Quine and those influenced by his paradigm are enjoining the ‘field linguist’ to depart radically from the procedures of the sciences, limiting themselves to a small part of the relevant evidence, selected in accordance with behaviorist dogma; and also reject the standard procedures used in theory construction in the sciences” (Chomsky, 2000, 54). According to this remark it is not justified to limit the evidence to suit one’s theory. I conclude that the Galilean style should be rejected.
Ariew, Roger. (1986). Descartes as Critic of Galileo’s Scientific Methodology, Synthese 67, 77–90.
Chomsky, Noam. (1957). Syntactic Structures. The Hague: Mouton.
Chomsky, Noam (1962) A Transformational Approach to Syntax. Third Texas Conference on Problems of Linguistic Analysis in English, Austin, Texas, The University of Texas.
Chomsky, Noam. (1966a). Cartesian Linguistics. New York: Haper&Row.
Chomsky, Noam (1966b) Topics in the Theory of Generative Grammar, The Hague, Mouton and Company.
Chomsky, Noam. (1972). Some empirical issues in the theory of transformational grammar. In: S. Peters (Ed.) Goals of linguistic theory. (pp. 63-130). Englewood Cliffs, NJ: Prentice Hall.
Chomsky, Noam (1975) Reflections on Language, New York, Pantheon Books.
Chomsky, Noam (1986a) Knowledge of Language, New York, Praeger Scientific.
Chomsky, Noam (1986b) Barriers, Cambridge, Massachusetts, The MIT Press.
Chomsky, Noam. (1995). The minimalist program. Cambridge, MA: MIT Press.
Chomsky, Noam. (2000). New Horizons in the Study of Language and Mind. Cambridge: Cambridge University Press.
Chomsky, Noam. (2002). On Nature and Language. Cambridge: Cambridge University Press.
Chomsky, Noam (2009). Opening Remarks. In: Piattelli-Palmarini, Massimo Uriagereka, Juan & Pello Salaburu (Eds.). Of minds and language: a dialogue with Noam Chomsky in the Basque Country. (pp. 13-43). Oxford: Oxford University Press.
Chomsky, Noam. (2012). The Science of Language. Cambridge: Cambridge University Press.
CSM I, II (1984/1986) The Philosophical Writings of Descartes, v. I, II, transl. John. Cottingham, Robert Stoothoff, & Dugald Murdoch. Cambridge: Cambridge University Press.
CSMK III. (1991). The Philosophical Writings of Descartes, v. III, transl. John Cottingham, Robert Stoothoff, Dugald Murdoch, & Anthony Kenny (Cambridge: Cambridge University Press.
Drake, Stillman. (1990). Galileo: Pioneer Scientist. Toronto: The University of Toronto Press.
Feyerabend, Paul. (1978). Redet nicht, organisiert Euch! In: Duerr, H. P. (Ed.) Unter dem Pflaster liegt der Strand, Bd. 10, (pp. 1-16). Frankfurt: Karin Kramer Verlag.
Fischer, Klaus. (1992). Die Wissenschaftstheorie Galileis – oder: Contra Feyerabend, Journal for General Philosophy of Science / Zeitschrift für allgemeine Wissenschaftstheorie, 23, 165-197.
Freidin, Robert. & Vergnaud, Roger. (2001). Exquisite connections: some remarks on the evolution of linguistic theory. Lingua 111, 639-666.
Kuhn, Thomas. (1996). The Structure of Scientific Revolutions: 50th Anniversary Edition. Chicago: University of Chicago Press.
McGilvray, James (2005.) The Cambridge Companion to Chomsky. Cambridge: Cambridge University Press.
How to cite this post
Behme, Christina. 2016. How Galilean is the ‘Galilean Method’?. History and Philosophy of the Language Sciences. https://hiphilangsci.net/2016/04/02/how-galilean-is-the-galilean-method