From WouterBeek.com

Contents 1 Introduction 2 Davidson’s argument in relation to the work of Kuhn 3 Kuhn’s refutation 4 Associating a language with a conceptual scheme 5 The syntax-nature link, or the problem of semantics 6 The stimulus-sensation route 7 Conclusion 8 Bibliography 9 References 10 Publication history

Introduction

Davidson, in his article ‘On the Very Idea of a Conceptual Scheme’^[1], gives an impressive argument in which he shows that talk about different world views, paradigms and points of view – all grouped under the header of ‘conceptual schemes’ – is entirely useless. For, so he argues, it will never be possible to establish whether such a variety of conceptual schemes indeed does exist, or whether there is only one such a conceptual scheme that is identical for all human beings. The concept of a conceptual scheme can thus not be rationally employed. Such an argument would of course invalidate the work of Thomas Kuhn, to whom the concept of incommensurability among revolutionary different scientific theories is a central notion. Kuhn has, in his article ‘Commensurability, Comparability, Communicability’^[2] argued that Davidson’s view of incommensurability is too strict, and that the argument does not hold with regard to Kuhn’s weaker notion of ‘local incommensurability’. I shall however show that such a defense is not undermining Davidson’s argument at all (and that such a remark has even been anticipated upon by Davidson in his article). I shall thereafter show that, even though Kuhn’s refutation is not to the point, there is a different reason why Davidson’s argument does not apply to Kuhn’s theory of scientific incommensurability, and that thus different paradigms may indeed still exist (and talk of them has not as of yet been shown to be irrational at all).

Davidson’s argument in relation to the work of Kuhn

Davidson in his article^[3] associates having a conceptual scheme with having a language.^[4] Different conceptual schemes must have different languages associated with them, and these languages are untranslatable into one another. At the same time different languages may share one and the same conceptual scheme, provided that these languages are translatable into one another.

Because Davidson’s paper is written in a generalized way, dealing with conceptual schemes in such a broad sense as to incorporate the views of – besides Kuhn – Feyerabend, Whorf, Quine and Bergson, our first task shall be to apply this general argument to the specific terminology and theory of Kuhn.

First we should make clear what is meant with the term ‘conceptual scheme’ with respect to the work of Kuhn. We may equate this expression with the term ‘paradigm’, but since this might give rise to misunderstandings – as has been the case ever since the term was coined by Kuhn in his Structure of Scientific Revolutions^[5] – we shall, in line with Kuhn’s later definitional refinement, make use of the term ‘disciplinary matrix’ instead.^[6] A disciplinary matrix constitutes the set of knowledge that is shared by the participants of a certain research group. This is a much broader use than the other meaning of ‘paradigm’, which was primary to the idea of Kuhn when he wrote Structure, but which has been secondary to his recipients (and which has sometimes been disregarded at all)^[7]. This more restricted sense of the term ‘paradigm’ designates the successful examples that show the applicability of a certain theory (i.e. paradigm in the broader sense) to a specific problem field. This restricted sense is, as we shall see below, a constituent part of the broader sense of ‘paradigm’. We shall – again following Kuhn’s reformulation subsequent to Structure – from now on use the term ‘exemplars’ to refer to this narrow sense of a paradigm.^[8]

A disciplinary matrix is the set of the different sorts of knowledge that are shared by the participants of a certain research group. This matrix of shared understanding is what accounts for the relative unanimity of both judgments and valuations that can be observed in a scientific community (and in almost no other sort of community). This matrix may thus be viewed as the characteristic attribute of a research community, the convention under which the efforts of its individual participants converge under the header of group achievement. The disciplinary matrix consists of, among other things, symbolic generalizations (or formulas), metaphysical and ontological commitments, shared values and – above all – the already introduced exemplars.^[9] To employ a disciplinary matrix is to view the world in a certain way, to have certain sensations when confronted with certain stimuli and to utilize a distinguished set of symbols with which to denote the ontological entities from whom those sensations are claimed to derive.^[10]

Kuhn’s refutation

Now, in reading Davidson, we understand his associating languages with conceptual schemes as associating languages with disciplinary matrices. It is to this association relation that Kuhn adheres also.

By accepting this identification relation between a language and a conceptual scheme, Kuhn has accepted the most fundamental assumption on the basis of which Davidson’s argument is based. So in Davidson we read:

“We may accept the doctrine that associates having a language with having a conceptual scheme. […] Studying the criteria of translation is therefore a way of focusing on criteria of identity for conceptual schemes.” [Davidson2001, p. 184]

This is echoed by Kuhn:

“The phrase ‘no common measure’ becomes ‘no common language’. The claim that two theories are incommensurable is then the claim that there is no language […] into which both theories, conceived as sets of sentences, can be translated.” [Kuhn2000, p. 36]

The argument of Davidson now shows, on the basis of the above assumption, that incommensurability of disciplinary matrices implies incomparability also. This however implies that there is no intelligible way in which different conceptual schemes might be calibrated. I shall not – for considerations of space – explicate Davidson’s exact argument here, but shall only describe Kuhn’s subsequent refutation of it.^[11]

Kuhn does not accept Davidson’s conclusion:

“No more in its metaphorical than its literal form does incommensurability imply incomparability […]” [Kuhn2000, p. 36]

The reason that Kuhn gives for his disagreement with Davidson is that the latter, so he claims, has attributed too strict a notion of incommensurability to him. Kuhn then coins a more modest version of incommensurability, which he calls ‘local incommensurability’, to describe his position. This more modest version states that even though in revolutionary shifts some terms are altered (some change their meaning, some become obsolete, some are newly introduced and even some parts of the ontology - to which the terms refer - change), all these changes are merely local, influencing only part of the total network of meaning, while the rest stays stationary. So the multitude of terms, together with the largest portion of the ontology, are claimed to remain stable in the process of revolutionary change.^[12]

This view seems to contradict some examples that Kuhn has provided on other occasions in which he points to the fundamental reform of a language that comes together with a revolutionary shift. So in ‘What are scientific revolutions?’^[13] he argues that Aristotle’s doctrine concerning the impossibility of a vacuum or void could not be altered by a falsifying experiment showing that such a vacuum indeed does exist. For allowing the idea of a void in Aristotle’s theory would have implied a redefinition of the notion of ‘location’ and of ‘natural place’, notions that both depend upon the inability of the existence of such a vacuum. Since these latter notions are central to his theory, changing them would necessarily imply changing many others as well. It is unclear what would remain of Aristotelian theory (and whether we would call it Aristotelian theory still) once these changes of definition would be propagated further along the lines of the interdependencies that exist between the various theoretical notions. Kuhn thus concludes:

“There is no way to ‘correct’ Aristotle’s views about the void without reconstructing much of the rest of his physics.” [Kuhn2001, p. 20]

So even though the idea of local incommensurability seems to conflict with Kuhn’s own theory of revolutionary change, let us grant that it indeed is the case that revolutions only involve some relatively minor and relatively local changes in the language employed by the scientist. Would this render Davidson’s argument obsolete? Kuhn suggests that the unchanged terms may provide a sufficient basis upon which the changed terms may be interrelated with their previous counterparts:

“The terms that preserve their meanings across a theory change provide […] a basis from which the meanings of incommensurable terms can be explored.” [Kuhn2001, p. 36]

In Davidson’s article however, no distinction is made between complete and partial failure of translatability. Because in both cases it is equally impossible to find a steady ground on the basis of which the existence of concepts radically different from one’s own may be suspected to exist (let alone be understood).^[14] Since Kuhn’s notion of local incommensurability can be seen as the equivalent of Davidson’s notion of partial failure of translation^[15], his introducing this notion does not invalidate Davidson’s argument at all.

Kuhn goes on to claim that interpretation and translation are not one and the same thing^[16], as is assumed by the theory of radical interpretation developed by Quine (which is employed by Davidson also). Even though translation may be impossible, i.e. a certain term in one language may not have a viable description in another, interpretation is still possible. For interpretation is not conceived of by Kuhn in the Quinian sense (and the two processes are not conflated). Kuhn provides the example of a text concerning the phlogiston theory. In it there are certain words, like ‘phlogiston’ or ‘principle’ that do not feature in the language of present chemical practice. Although these notions can be translated into the present language to some extent, they can only show that the phlogiston theory once worked, i.e. they may explain what the term ‘phlogiston’ refers to in a certain situation, but they cannot guide phlogiston research with regard to new problem instances.^[17]

But even though translation of the phlogiston text may be impossible, interpretation of the text in always possible. Such an interpretation would require a person to learn the language of phlogiston theory from the ground up. It will not be taught (at least not entirely) through the present chemical language, but will be taught alongside, i.e. separated from it. Although the two may share some terms and some meanings (according to the notion of local incommensurability), they cannot be entirely interrelated. And a person switching from one language to the other would be making a switch of Gestalt. His research in the one language cannot be fully related to his research in the other (and vice versa). But this is just the same thing as stating that there is no intelligible way in which to relate expressions from the one language to expressions in the other. This statement alone need of course not imply that talking about conceptual schemes is irrational too, but combined with the tight association of a language with a conceptual scheme, it does. This line of reasoning thus fails to defy Davidson’s argument too.

Associating a language with a conceptual scheme

Now I shall set out my own views regarding the applicability of Davidson’s argument to the work of Kuhn. I shall therefore start out with inquiring how a disciplinary matrix and a language can be associated. For some constituents of the disciplinary matrix this association is unproblematic, because some parts of it are already available in a linguistic form. We have the set of terms and symbols to be employed (e.g. ‘mass’, ‘electron’, ‘f’, ‘Σ’), which constitute the vocabulary of our language. Furthermore we have the formulas and generalizations that specify the relations that must hold among these terms and symbols (e.g. 'f = ma', ‘there can exist no vacuum’). The latter constitute – at least partially – the syntactical operations (both mathematical and logical) that can be applied to these symbols. These may be called – again at least partially – the syntax of our language. In conjunction to that we have the ontology which derives from the metaphysical and ontological commitments, which describes the entities that exist in the world and that are relevant from a scientific point of view (e.g. electrons, phlogiston, vacuum, gravitational attraction). Let it be clear that the language thus obtained still remains underspecified by these notions (and this is why I have employed the phrase ‘at least partially’ in the above). But what is important, is that the space of possible languages that associate to a disciplinary matrix in this way is already quite narrow. Although there may be several versions possible – all of which show some deviations with respect to one another – they will be fairly similar, and provide relatively little communicational difficulties. We may now assume to have a relatively comprehensive description of both the syntactical component of our language, as well as a model of nature. Both are associated with our conceptual scheme/disciplinary matrix. But not all has been done yet, for there is still a crucial flaw in the thus derived language. The way in which the two described components – i.e. the syntactical and the ontological – interrelate is still entirely unspecified. For to know the vocabulary of a language and to know the manipulatory rules that apply to them on the one hand, and to know the ontology of elements and events in nature on the other, is not enough. We need to interrelate these two components. For even though there may be agreement over the application of logical and mathematical manipulations upon certain strings, and even though there may be some agreement over what entities and events the world consists of, it is still an open question how these symbols relate to the concrete observations made and to the concrete experimental results obtained. Kuhn is still with us on this one:

“A shared commitment to a set of generalizations justifies logical and mathematical manipulation and induces commitment to the result. It need not, however, imply agreement about the manner in which the symbols, individually and collectively, are to be correlated with the results of experiment and observation.” [Kuhn1977, p. 299]

The syntax-nature link, or the problem of semantics

This problem has been called the ‘language-nature link’ by Kuhn.^[18] But I would like to replace it with a term that I think better fits the problem at hand. As we have seen in the above, the linguistic portion that could easily be derived from our conceptual scheme was the syntactical component only. It then becomes clear that the term ‘language’ in ‘language‐nature link’ refers to the syntactical component of a language. (This refinement is not as trivial as it may seem, for the term ‘language’ ordinarily refers to both the syntactical and the semantic component.) This latter module, however, is exactly what we signify when we use the phrase ‘language‐nature link’. So we may elucidate things a bit further by replacing the ambiguous term ‘language‐nature link’ by ‘syntax-nature link’. Renaming the problem has of course not solved it, although it has provided us with a more acute terminological apparatus to further the elaboration of this problem.

Let us now focus upon this flaw in our present language, which we will now call the ‘syntax-nature link’, or – which is equivalent – the problem of semantics. There is a portion of the disciplinary matrix that relates to the problem of semantics, and it has been widely described by Kuhn. It is the set of exemplars (i.e. paradigms in the narrow sense) already alluded to in the above. These exemplary solutions to particular problem instances illustrate the way in which the formulas and generalizations (i.e. the syntax of the language) are to be related to the entities that have been postulated by the ontological and metaphysical presuppositions (i.e. the notion of nature). This linking of the two cannot be established by again providing another rule, for rules interrelate symbols to symbols, and applying them will thus never lead us beyond the syntactical component of a language. In order to cross the divide between symbol and reality something else is needed, something that – in line with the work of the latter Wittgenstein – may be called the assignment of meaning to a term (or an expression) through the term’s (or the expression’s) concrete use in the world.^[19] It is through this practical usage – application to concrete problem instances, that is – that the scientist comes to identify new research problems (and their subsequent solutions) as like the exemplars within his or her disciplinary matrix.^[20] This recognition of a similarity relation is prior to any form of verbalized knowledge regarding what it exactly is that this similarity relation consists of (this explicit knowledge is signified by Kuhn as ‘correspondence rules’):

“[…] an acquired ability to see resemblances between apparently disparate problems plays in the sciences a significant part of the role usually attributed to correspondence rules.” [Kuhn1977, p. 306]

And:

“To a man who speaks of similarity or of analogy, we […] at once pose the question: similar with respect to what? In this case, however, that is just the question that must not be asked, for an answer would at once provide us with correspondence rules.” [Kuhn1977, p. 307]

But we now begin to wonder in what way a disciplinary matrix could ever be fully stated in a language, any language at all. For how should we, in a single language that is, give a full description of the set of exemplars? Such would seem to be impossible, there will be no language at all capable of describing the full structure of the cognitive apparatus in this respect. To assume that there would be a language that would fit a disciplinary matrix is to assume that such a disciplinary matrix could be entirely verbalized.

But, as we have already seen, although some of its components do have a linguistic bearing (some are already given in a linguistic format, e.g. the symbols, the formulas, the generalizations), there are also some components that are in principle unverbalizable (e.g. the exemplars).

The stimulus-sensation route

Kuhn seems to find some correspondence between his ideas and those of Benjamin Lee Whorf. The essential thesis is Whorf’s work is that the language a person employs shapes his or her mind in a certain way. It alerts the person to different aspects in his or her environment. The speakers of different languages thus come to observe different things. Now this has an important overlap with the ideas of Kuhn as expressed in the postscript of the second (and third) edition of Structure, where he refers to what he calls the ‘stimulus-sensation route’.^[21] This term refers to the neural process that has been shaped by a long period of education, i.e. by being exposed to certain exemplars whereas not to others.

The thus shaped cognitive apparatus is able to process the stimuli that enter the sense organs of a person and construct more complex sensations out of them. The way in which this stimulus-sensation route processes and composes the data it receives depends heavily upon the sort of education that a person underwent and the sort of problem solutions that he or she has cognized. Kuhn gives numerous examples of changes of perception in the history of science.^[22] He draws an analogy between these changes of perception and the well-established results in experimental psychology, in which prior exposure to certain stimuli has been show to influence a person’s perception at subsequent stages. This process of composing complex sensations out of simple stimuli is to the person him- or herself entirely inaccessible:

“We have no direct access to what it is we know, no rules or generalizations with which to express this knowledge.” [Kuhn1996, p. 196]

We may thus single out the topic of the stimulus-sensation route as the exemplary part of cognition to which Kuhn’s argument of incommensurability among revolutionary different disciplinary matrices applies. There may be some other areas besides, but they are higher up in the hierarchy of cognition and their systems are more likely to be conflated with accessible, explicitly verbalizable knowledge, which would obscure the case and hamper the search for the so-called ‘tacit knowledge’ that we are after. We may thus assume that if Kuhn’s notion of tacit knowledge can be found anywhere at all, it must surely be present in the stimulus-sensation route.

Let us focus upon what a scientist does when practicing science. The scientist will of course never be in direct contact with reality, but will instead be in direct contact with a set of stimuli. Now to do science means to employ the symbols that are given with the disciplinary matrix. And to employ them is to employ them onto some entities or events that are to be found in the world. The kind of entities and events that are to be found are grouped in the ontology of the disciplinary matrix used. To employ the symbols is to attach them in some way or another to certain entities or events that feature in the ontology. The mathematical and logical manipulations that are applied to the symbols will, because of the attachment operation, be regarded as being applied to nature itself, and will thus constitute knowledge of the world, through the manipulation of symbols only. There is however one further complication that needs to be addressed, namely that sensations are not basic and given in the world but are already complex and construed in the mind of the individual scientist. So there are two difficult tasks that the scientist has to undertake. In the first place he or she has to construe sensations out of stimuli. In the second place he or she has to attach symbols to these complex sensations.

According to Kuhn both of these processes can best be described as linguistic processes. But I do not agree. For even though the second task may arguably be linguistic, the first one surely is not. For even though the construction of complex sensations from simple stimuli may in some sense be guided by the language one uses, this surely need not be the case. The stimulus-sensation route is adapted by a person's past experience. But this experience is only partially linguistic. Kuhn himself repeatedly provides us with examples from Gestalt psychology, which invariably involve visual conditioning of the mind. So when a person has been exposed to ducks a lot, he or she will be more likely to see a duck, less likely to see a rabbit. But such examples cannot be provided for language. The example of differences in color perception due to differences in color vocabularies^[23] has been sufficiently disproved.^[24] Although claims concerning linguistic influences upon the cognitive apparatus have been made by, among others, Sapir and Whorf, they have almost never been made with regard to the stimulus-sensation route. It should thus be equivocal that linguistic influences are not the primary cause in shaping the stimulus-sensation route. Such influences have, however, been show for nonlinguistic factors, most notably visual experience (but also experiences regarding other sense-organs). As for language, which is not an experience of any one of the sense-organs, it is very unlikely to have any substantial influence (if any influence at all) upon the stimulus-sensation route.^[25]

Conclusion

So what I would like to argue is that, as Kuhn claims, a new disciplinary matrix does indeed involve a reestablishment of the attachment of symbols to sensations, and a reestablishment of the way in which these sensations are derived from stimuli. Even the set of symbols employed and the set of sensations constructed need not be (entirely) the same, which gives the process of scientific revolution its dramatic effect that cannot be resolved be translation (in any usual sense of the word). These processes are however not entirely linguistic, but consist (at least partially) of alternations to the cognitive apparatus in a way that may most nearly be described by the term ‘conditioning’.

This exposition would perhaps at first seem to be quite unrelated to the above announced treatment of Davidson’s critique, but as a matter of fact the two do relate. For in ending to call the problem of employing a disciplinary matrix an entirely linguistic problem, we have violated Davidson’s fundamental assumption on which his argument bears, namely the association of a disciplinary matrix with a language. Of course there is some association, for some of the components of the disciplinary matrix can come to (more or less) determine a language. But some other components, most notably the exemplars of successful problem solutions, do not. They do provide the link between the syntactically specified language and the concrete new problem instances that are approached by the scientist, but are essentially non-linguistic. Violating the doctrine of associating a language with a disciplinary matrix results, according to Davidson, into the following problem:

“If conceptual schemes aren’t associated with languages […], the original problem is needlessly doubled, for then we would have to imagine the mind with its ordinary categories, operating with a language with its organizing structure.” [Davidson2001, p. 184]

But this need only be so if the mind and the language are both competing over the same cognitive tasks. This is not necessarily the case, for some portions of cognition may be explicit and structured by language (notably the higher cognitive functions, generalizations, abstractions). Other portions of cognition are implicit (or tacit) and are structured by the neural apparatus in a way that is principally unverbalizable (especially the lower cognitive functions, e.g. the stimulus-sensation route).

By accepting this ‘doubling of the problem’, we have invalidated Davidson’s critique – however useful it may be for those who do adhere to the assumptions he posits – with regard to the process that Kuhn’s scientist is engaged in.

Bibliography

Davidson, Donald. 2001. ‘On the Very Idea of a Conceptual Scheme.’ In: Inquiries into Truth and Interpretation. 2nd edition. Clarendon Press: Oxford.

Heider, E.R. 1972. 'Universals in color naming and memory.' In: Journal of Experimental Psychology, Vol. 93, p. 10-20.

Kuhn, Thomas S. 1977. The Essential Tension. Selected Studies in Scientific Tradition and Change. The University of Chicago Press: Chicago and London.

Kuhn, Thomas S. 1996. The structure of Scientific Revolutions. 3rd edition. The University of Chicago Press: Chicago and London.

Kuhn, Thomas S. 2000. The Road Since Structure. Philosophical Essays, 1970-1993, with an Autobiographical Interview. Edited by James Conant and John Haugeland. The University of Chicago Press: Chicago and London.

Wittgenstein, Ludwig. 1988. ‘Philosophische Untersuchungen.’ In: Werkausgabe, Band 1. Suhrkamp.

References

1. ↑ Davidson2001, p. 183‐198.
2. ↑ Kuhn2000, p. 33‐57.
3. ↑ Again ‘On the Very Idea of a Conceptual Scheme’ In: Davidson2001, p. 183‐198.
4. ↑ Davidson2001, p. 184.
5. ↑ For Kuhn’s regret in making too broad a use of the term ‘paradigm’, see the following citation:

   “Unfortunately […] paradigms took on a life of their own […]. Having begun simply as exemplary problem solutions, they explained their empire to include, first, the classic books in which these accepted examples initially appeared and, finally, the entire global set of commitments shared by the members of a particular scientific community.” [Kuhn1977, p. xix]

   And Kuhn’s exclamation in an interview he gave late in life shows a similar regret:

   “Paradigm was a perfectly good word, until I messed it up.” [Kuhn2000, p. 298]

   It is a little ironic that exactly this term, that was later regretted and abandoned by its originator, would become the very epitome of Structure and of Kuhn’s work in general.
6. ↑ See ‘Second Thoughts on Paradigms.’ In: Kuhn1977, p. 297.
7. ↑

   “The paradigm as shared example is the central element of what I now take to be the most novel and least understood aspect of this book [=Structure].” [Kuhn1996, p. 187]

8. ↑ See ‘Postscript.’ In: Kuhn1996, p. 187.
9. ↑ A precise description of a disciplinary matrix is never given by Kuhn. The incomplete list of constituents given has been derived from both the ‘Postscript’, in Kuhn1996, p. 182‐187 and from ‘Second thoughts on paradigms’, in Kuhn1977, p. 297‐298.
10. ↑ Such has been the central claim of Structure (Kuhn1996), although the term ‘paradigm’ is used here instead of the term ‘disciplinary matrix’. See especially the tenth chapter: ‘Revolutions as Changes of World View’, p. 111‐135, in which difference of paradigm is linked to difference of world view.
11. ↑ Davidson’s argument is quite brilliant and quite concise in its own respect. Summarizing it would be almost impossible.
12. ↑

    “Most of the terms common to the two theories function the same way in both; their meanings […] are preserved […]” [Kuhn2001, p. 36]

13. ↑ The same point, along with ingenious examples of both Volta’s cell and Planck’s work on the black‐body problem, also feature in the –still unpublished – second Thalheimer lecture.
14. ↑

    “We must conclude […] that the attempt to give a solid meaning to the idea of conceptual relativism, and hence to the idea of a conceptual scheme, fares no better when based on partial failure of translation than when based on total failure.” [Davidson2001, p. 197]

15. ↑ The notion of partial failure of translation is defined as follows: “[…] there would be partial failure if some range [of sentences] could be translated and some range could not […]” [Davidson2001, p. 185], which seems to match Kuhn’s definition of local incommensurability.
16. ↑ This line of argument is provided in Kuhn2000, p. 44‐45.
17. ↑ In Structure Kuhn had already provided a similar example of translating a prior theory in terms of a latter one:

    “[…] even if that transformation [=stating Newtonian theory in terms of Einsteinian theory] were a legitimate device to employ in interpreting the older theory, the result of its application would be a theory so restricted that it could only restate what was already known. […] it could not suffice for the guidance of research.” [Kuhn1996, p. 103]

    For a more elaborate explanation of this, see the entire section from which this citation was taken, in Kuhn1996, p. 101‐103.
18. ↑ Kuhn1977, p. 303, footnote 13.
19. ↑

    “Was bezeichnen nun die Wörter dieser Sprache? – Was sie bezeichnen, wie soll ich das zeigen, es sei den in der Art ihres Gebrauchs?” [Wittgenstein1988, p. 242]

20. ↑ Kuhn often alludes to Wittgenstein’s notion of family resemblance when explaining the recognition of the similarity of new problem instances to the exemplars within the disciplinary matrix. See, for example, Kuhn1996, p. 45.
21. ↑ Kuhn1996, p. 196.
22. ↑ For example:

    “During the seventeenth century, when their research was guided by one or another effluvium theory, electricians repeatedly saw chaff particles rebound from, or fall off, the electrified bodies that had attracted them. At least that is what seventeenth‐century observers said they saw, and we have no more reason to doubt their reports of perception than our own. Placed before the same apparatus, a modern observer would see electrostatic repulsion (rather that mechanical or gravitational rebounding) […]” [Kuhn1996, p. 117]

    Another example is the different perception Volta had of the battery (already alluded to in the above), Kuhn2000, p. 20‐24.
23. ↑ This example has occasionally been alluded to by Kuhn, for example in Kuhn1977, p. 308‐309.
24. ↑ See especially Heider1972.
25. ↑ This is exactly the reason why Sapir and Whorf have at several occasions pointed out that the most important influences of language upon thought are to be found within the higher cognitive functions (where language resides).

Publication history

This article was written for a course in philosophy of science, taught by Prof. Dr. G. de Vries, sometime in January 2007. It has been slightly altered since.