This essay was written for a class, History of Analytic Philosophy Part II, at University of California, Irvine. The course was taught by Professor Jeremy Heis in the Spring Quarter of 2019.

I am quite proud of this essay. However, it is missing some coverage of a few topics, which I list below at the bottom of this post in the section Further Areas for Elaboration. If I seriously revisit this essay, I will add discussions about those areas.

Fruitful Explication in Philosophical Investigations

The Vienna Circle and Carnap take philosophy to play an active role with science in several ways.  They generally hold that philosophy can help unify the sciences, make our language (more) scientific/exact, and coordinate with science to make fruitful advancements.  In contrast, Wittgenstein’s view of philosophy in §§109-133 of Philosophical Investigations (1953) [PI] famously describes that “the real discovery is the one ... that gives philosophy peace” and that philosophical problems are in need of methods comparable to “therapies”.[1]  The therapy analogy is akin to how therapists would not want to legitimate a paranoid man’s worries by appeasing each worry, and would instead seek to dissolve his worries altogether by dissolving the source of the paranoia.  Wittgenstein thinks philosophical problems should not be taken as legitimate problems in themselves (e.g. discovering the true meaning of ‘justice’), and should be addressed through methods which put to rest whatever has brought the philosophical problems into question in the first place (e.g. there is nothing more to the meaning of ‘justice’ than the way we use the word).  In that way, Wittgenstein thinks that philosophical “[p]roblems are solved (difficulties eliminated), not a single problem.”[2]  It seems like the Vienna Circle’s and Carnap’s view that philosophy has a fruitful role in coordination with science might be incompatible with Wittgenstein’s view that it is best to give philosophy peace.

I believe that philosophy has the potential to be fruitful alongside science, but I agree with Wittgenstein’s description of philosophical problems in PI as resulting from philosophers’ misunderstandings of language—I don’t think that these views are as incompatible as I implied them to be in the previous paragraph.  I will first describe the similar anti-metaphysical stances among the Vienna Circle, Carnap, and Wittgenstein, and how their views diverge based on their different views of the possibility of a logical/scientific language.  Then, I focus on Carnap’s views as they are in his chapter “On Explication” [OE] in Logical Foundations of Probability (1950), which I read to be almost compatible with the discussion of language and the role of philosophy in PI.  Carnap’s views in OE, if read very generously, slot quite neatly into a space which, while recognized by Wittgenstein in PI, is left mostly unexplored (at least in PI §§1-133).[3]  I tend to find later-Wittgenstein to be more convincing than Carnap and the Vienna Circle where their views disagree about metaphysics and language.  Yet as I am sympathetic to the idea that a thoughtful coordination of philosophy and science can be fruitful, I am glad to see some compatibility among later-Wittgenstein and Carnap’s OE views.

The Vienna Circle’s and Carnap’s Anti-Metaphysical Stance

The Vienna Circle, Carnap in Logical Syntax, and Wittgenstein in Philosophical Investigations all hold anti-metaphysical views, but in slightly different ways.  How they describe their issue with metaphysics is linked to what they think the role of philosophy should be.  In this section, I focus on the views of the Vienna Circle, and Carnap in The Logical Syntax of Language.

The Vienna Circle, in The Scientific Conception of the World (1929) [TSCW], reject metaphysical sentences because they think that knowledge only comes “from experience, which rests on what is immediately given” or else is a tautological transformation which adds no more content to what there was before.[4]  Under this view, metaphysical statements—which try to say something about the world a priori—would then only have epistemic content insofar as they are reducible to empirical statements.  Apart from that, the Vienna Circle thinks metaphysics should be discarded from science and philosophy.  The goal of philosophy according to TSCW is to use logical analysis to reduce scientific statements into statements about given sense-data, which would both unify the sciences through the language of sense-data and would make science accessible to everyone because everyone is presumed to understand the language of sense-data.[5]

~

Carnap, in The Logical Syntax of Language (1934) [LS], updates his views about the separation between logical and empirical sentences.  Carnap is now tolerant to which formal qualities in their languages people choose to use, instead of claiming that there is one best universal language of sense-data.[6]  This means that people are free to choose what rules in logic and axioms in mathematics they want to hold, which is different from the idea that logic is only the Tractatus-inspired tautological transformations which the Vienna Circle believed in.  Carnap distinguishes between first-order ‘object’ sentences and second-order ‘syntactical’ sentences, which respectively “speak either in or about the sentences of [a scientific domain].”[7]  Traditional philosophical questions turn out to be psychological questions (e.g. for epistemology), expressions of feelings (e.g. for normative sentences), syntactical questions about the scientific language (for questions about objects found in the sciences, e.g. time, space, society), and/or are found to mistakenly treat relative syntactical structures that are contingent on the language used (e.g. numbers, metaphysics) as absolute objects to be studied.[8]  As it was with the Vienna Circle, metaphysical and non-scientific questions are denied any epistemic import.

According to LS, the role of traditional scientists is mostly to use first-order object sentences to write scientific papers/treatises.  The role of traditional philosophers is shifted to using second-order sentences to talk about the “logical, formal, syntactical connections” within scientific language.[9]  However, scientific treatises will still have to use some syntactical statements, and philosophical (logic of science) writings likewise will still use some object statements.  Philosophers and scientists are considered to be working in the same field, albeit with “differences of degree” in how much they focus on objects versus on syntax.[10]  Because they work in the same field, Carnap’s hope in LS is that scientists and philosophers will be able to engage in “fruitful, co-operative work.”[11]  Carnap thinks metaphysical, non-scientific questions are creating confusion and thus getting in the way of such fruitful cooperation, which is why he seems to want people to reject metaphysics in LS.

Both TSCW and LS focus on philosophy’s role of disambiguating the language around science.  In TSCW, this means the “clarification of problems and assertions.”[12]  In LS, studies into syntax will mean that we will “have as our subject-matter exact terms and theses that can be clearly apprehended.”[13]  Both TSCW and LS take this clarification of language to be achievable—either by translating scientific language into the language of sense-data or by clearly stating the scientific language’s syntactic formation and transformation rules.  This view about the clarification of language seems to be inspired by advancements in science which the Vienna Circle and Carnap took to come from rejections of metaphysical language, such as Einstein’s theories of relativity rejecting absolute time and space, or theories in biology that reject the vitalist theory of a “special life force.”[14]  They were also primed by Wittgenstein’s Tractatus Logico-Philosophicus and Russell’s Our Knowledge of the External World (both cited by the Vienna Circle as representative of their thought) to think that there could be a logical language in which scientific thought could reach exactness.

Wittgenstein’s Anti-Metaphysical Stance

Wittgenstein, in Philosophical Investigations, agrees that metaphysical questions are troubled because of mistakes about language.  However, Wittgenstein’s description of language differs from the Vienna Circle’s and Carnap’s views in TSCW and LS.  Particularly, Wittgenstein seems to disagree about the exactness achievable in language.  Consequently, we might think that he would disagree with the conclusion in LS that philosophy has the potential to be in fruitful cooperation with science.  Indeed, with Wittgenstein saying that philosophy should be given peace (as described in the first paragraph of the essay), it seems as though Wittgenstein is saying that philosophy is best not being done at all.

Wittgenstein is against metaphysics because he thinks that language is a tool used by people to coordinate.  There are situations where words become vague—that is, it is ambiguous as to whether or not a word applies—because we haven’t had to describe/coordinate around those certain situations before.  For example, when does something become ‘art’?  Or when is something no longer considered a ‘chair’?[15]  In these vague situations, we are prone to misunderstand language such that we think there’s a true meaning behind the vagueness.

Wittgenstein seems to think it wrong-headed “[w]hen philosophers use a word—'knowledge’, ‘being’, ‘object’...—and try to grasp the essence of the thing”.[16]  Instead of engaging in such metaphysics, “we ... bring words back from their metaphysical to their everyday use.”[17]  There is no true meaning of a word that could cut through the vagueness—instead, in most cases, the meaning of a word is merely how it is used every-day, ambiguities and all.

Thus, to Wittgenstein in PI, philosophy is about description, not explanation of the words we use, and we need to use methods to give philosophy peace wherever people feel the urge to pin down metaphysical essences.[18]  The purpose of philosophy is to make language “surveyable” so we can see where the “bumps” in our language are—that is, which situations our language is not useful in.[19]  Philosophy “must not interfere in any way with the actual use of language”; the best we can do is intervene before people get lost down metaphysical rabbit-holes.[20]

Disagreements About the Possibility of Logical Language

Wittgenstein is anti-metaphysical, similar to the Vienna Circle and Carnap in LS, but he disagrees with them about the possibility of a logical language which might disambiguate metaphysical problems.  To Wittgenstein, a ‘language of sense-data’ or ‘syntactical formation and transformation rules’ would be misguided in a similar way as metaphysics—such projects presuppose an essential logic behind language.

Carnap and the Vienna Circle might look at advancements in science by Einstein and others as examples of how a non-metaphysical stance has the capacity to produce exactness in science.  But Wittgenstein would think that this is a mistake: “we eliminate misunderstandings by making our expressions more exact; but now it may look as if we were aiming at a particular state, a state of complete exactness, as if this were the real goal of our investigation.”[21] Successful changes in language which make language more usable do not mean that there is a final logical language that can express complete exactness.

Putting Carnap’s OE into Compatibility with Wittgenstein’s PI

In “On Explication” [OE], a chapter in Logical Foundations of Probability (1950), Carnap further describes how philosophy can be fruitful through the process of ‘explication’: the replacement of unscientific concepts with better concepts, each of which is similar to the concept it replaces, exact, fruitful, and then as simple as possible.[22]  I worry that Carnap is still making the same mistake of requiring exactness, but he describes that ‘exactness’ serves to “introduce the [new term] into a well-connected system of scientific concepts.”[23]  Insofar as ‘exactness’ simply means ‘integration with existing scientific language’, Carnap’s views in OE seem to slot nicely into a space in PI.

I should note that Carnap in OE still likely believes in an exactness achievable through logic, given how he talks about axioms and Frege’s arithmetic at the end of the chapter.[24]  However, it is worth remembering that Carnap’s tolerance to different languages in LS indicates that he is not hard-set on any one language—just whichever works best for the science.  That, at least, matches Wittgenstein’s view that the shape of our language tends to follow our use cases.

~

Wittgenstein praises the goal of a well-connected language while discussing the role of philosophy as creating a survey of our language: “A surveyable representation produces precisely that kind of understanding which consists in ‘seeing connections’.  Hence the importance of finding and inventing intermediate links.”[25]  If explications provide more “exact” concepts which are “well-connected” to the scientific language, then Carnap is seemingly describing the creation of intermediate links that Wittgenstein is there praising.[26]

Wittgenstein is open to solving misunderstandings in our language not simply by therapeutic methods, but also through something that is in many ways similar to Carnap’s process of ‘explication’.  Similar to how Carnap in OE advocates the replacement of parts of language, Wittgenstein says that “some [misunderstandings concerning the use of words] can be removed by substituting one form of expression for another; this may be called ‘analysing’....”[27]  Wittgenstein’s ‘analysis’ seems to be a valid way of fixing some of the misunderstanding-prone parts of our language.  It replaces problematic vagueness with unproblematic language, like how we might imagine an explicated concept to be more useful to science than an unscientific unexplicated concept.

Then why does Wittgenstein still think that it would be good for philosophy to find peace?  First, therapeutic methods solve problems, “not a single problem”, which is the capacity of analysis.[28]  But more importantly, I think the famous section §133 about “[giving] philosophy peace” needs to be read in the context of the section that precedes it.[29]  In that preceding section, Wittgenstein writes:

[A] reform for particular practical purposes, an improvement in our terminology designed to prevent misunderstandings in practice, may well be possible.  But these are not the cases we are dealing with.  The confusions which occupy us arise when language is, as it were, idling, not when it is doing work.[30]

When Wittgenstein speaks of giving philosophy peace, he is putting aside cases of language changing in practice—he cares about the philosophical problems which arise without provocation, when people misunderstand language to have an essence.  Therefore, when Carnap describes making explications, Wittgenstein might see him as describing ‘analyses’ that prevent misunderstandings in cases in which language is doing work—scientific work.  The prevention of misunderstandings and the creation of intermediate links puts science into a better position to make universal statements, and, at least in “On Explication”, Carnap sees that as ‘fruitful’.[31]

Further Areas for Elaboration

I can see a few areas of this essay which would benefit from more elaboration:

1. Wittgenstein’s view on exactness is that it is always in respect to a goal, and that there doesn’t seem to be a satisfactory way of saying what the proper goal is. Particularly, in section §88, he describes that “… ‘Inexact’ is really a reproach, and ‘exact’ is praise. And that is to say that what is inexact attains a goal less perfectly than does what is more exact. So it all depends on what we call ‘the goal’.” What goals does Wittgenstein seem to value (especially in §§109-133, where he lays out his ideas on the role of philosophy)?

2. What I call Wittgenstein’s ‘analysis’ in this paper is more likely something which originates with Russell. The Vienna Circle also appreciated how Russell’s Our Knowledge of the External World analyzed complex concepts like ‘space’ and ‘time’ as complex logical constructions of more basic components of sense data. It would be good to recognize this in the essay, even if it is just in a footnote.

3. About the last line of the essay, why does preventing misunderstandings and creating intermediate links mean we are more capable of making universal statements?

4. What is the consensus contemporary interpretation of Wittgenstein’s “give philosophy peace” statement?

5. What does Wittgenstein think are good goals around developing language, given that he thinks that we should seek to describe, not explain, language? What are the consequences for conceptual analysis and philosophy’s role with science? For example, Wittgenstein scoffs at attempts to define knowledge, but isn’t that what scientists are often attempting to expand? What are the limits of analysis, and are there ambiguities in the outline that Wittgenstein provides for when we are using language to do work versus getting caught up in the mess of ambiguous language?

6. Carnap’s Logical Syntax might help us with the last question in (5). We want to be able to develop a role for philosophers to aid science, but Wittgenstein points out that some attempts to disambiguate words are given too much importance due to a misunderstanding about language (and deserve less emphasis, thought, and supposed importance). In the essay, I may have been too harsh on Carnap’s LS. There, he doesn’t think that there can be a perfectly logical language, but says that he wants to focus on the ways in which it seems to make sense for language to be interpreted as a logical system. This is seemingly an admission that language is sometimes not possible to fit into a logical language, but that, at least in many cases of science/mathematics, there are considerable benefits to understanding language to have a certain logic of formation and transformation rules.

Footnotes

[1] Ludwig Wittgenstein, Philosophical Investigations (Chichester, West Sussex: Wiley-Blackwell, 2009), §§133, 133d.

[2] Ibid., §133.

[3] Ibid., §132.

[4] Rudolf Carnap, Hans Hahn, and Otto Neurath. “Wissenschaftliche Weltauffassung: Der Wiener Kreis [The Scientific Conception of the World: The Vienna Circle],” in Empiricism and Sociology, ed. Marie Neurath and Robert S. Cohen (Doedrecht, Holland: D. Reidel Publishing, 1973), 299-318.

[5] Ibid., 309.

[6] Carnap, Rudolf. The Logical Syntax of Language (Chicago: Open Court, 1934), 51.

[7] Ibid., 277, 331.

[8] Ibid., 276-277, 298-299.

[9] Ibid., 332.

[10] Ibid., 331.

[11] Ibid., 333.

[12] Carnap, Hahn, and Neurath, “The Scientific Conception of the World,” 306.

[13] Carnap. The Logical Syntax of Language, 333.

[14] Carnap, Hahn, and Neurath, “The Scientific Conception of the World,” 310-315.

[15] Wittgenstein, Philosophical Investigations, §80.

[16] Ibid., §116.

[17] Ibid., §116

[18] Ibid., §§109, 133.

[19] Ibid., §§126, 119.

[20] Ibid., §124.

[21] Ibid., §91.

[22] Rudolf Carnap, “On Explication,” in Logical Foundations of Probability (Chicago: University of Chicago Press, 1950), 7.

[23] Ibid.

[24] Ibid., 17-18.

[25] Wittgenstein, Philosophical Investigations, §122.

[26] Carnap, “On Explication,” 7.

[27] Wittgenstein, Philosophical Investigations, §90.

[28] Ibid., §133.

[29] Ibid.

[30] Ibid., §132.

[31] Carnap, “On Explication,” 7.

Bibliography

Carnap, Rudolf. The Logical Syntax of Language. Chicago: Open Court, 1934.

Carnap, Rudolf. “On Explication.” In Logical Foundations of Probability, 1-18. Chicago: University of Chicago Press, 1950.

Carnap, Rudolf, Hans Hahn, and Otto Neurath. “Wissenschaftliche Weltauffassung: Der Wiener Kreis [The Scientific Conception of the World: The Vienna Circle].” (1929) In Empiricism and Sociology, edited by Marie Neurath and Robert S. Cohen, 299-318. Doedrecht, Holland: D. Reidel Publishing, 1973.

Wittgenstein, Ludwig. Philosophical Investigations. With translation by G. E. M. Anscombe, P. M. S. Hacker and Joachim Schulte. 4th ed. Chichester, West Sussex: Wiley-Blackwell, 2009.