Wittgenstein and Grounding: How Language Creates
Shauna Blake 11782862 August 2018
Table of Contents
1.0 Introduction - - - page 2 2.0 Wittgenstein’s theory of meaning as use in the Investigations - - - page 3 2.1 Linguistic idealism - - - - page 7 2.2 Some criticism with responses - - - page 14 2.3 Some advantages to linguistic idealism - - - page 17 3.0 Introduction to grounding - - - page 18 3.1 Kit Fine’s grounding- - - page 25 3.2 Bringing together grounding and linguistic idealism - - - - page 27 4.0 Conclusion - - - - page 33 Bibliography - - - - page 35
1.0 Introduction
I will be presenting Wittgenstein’s linguistic idealism as a grounding thesis. Though since it isn’t the conventional reading of him I’ll first be establishing how Wittgenstein can be read as a linguistic idealist. The central claim of linguistic idealism is that the objects of our world are dependent on language. There are some issues with Wittgenstein’s linguistic idealism surrounding clarity and mechanics that I think can be addressed with grounding. In particular Kit Fine’s grounding, which preserves the flexibility that is an important element of Wittgenstein’s use of the entirety of everyday language. The combination Fine’s grounding and Wittgenstein’s linguistic idealism, leads to this: an object is the object it is in virtue of the linguistic practices that invoke it. How an object occurs in language and is treated by language users grounds the object being what it is. To show this in chapter 2 I will start with examining the late Wittgenstein and how he can be read as a linguistic idealist, and what that entails. In chapter 3 I will then turn to grounding, and how it addresses the weaknesses of linguistic idealism.
Wittgenstein is notoriously opaque, as such his writing tends to be read in many different ways, a more controversial one of which is as a linguistic idealist. Linguistic idealism is not particularly popular, one reason for that is how easily misunderstood it can be. The central claim of linguistic idealism is that our language shapes the reality of our world. Because of that exactly what is meant by “language” is a central issue here, especially since it has been understood in dramatically different ways over the years. But once it’s established what is meant by each of the parts linguistic idealism, avoiding many of the problems many people have with linguistic idealism stemming from their views on language, it becomes much more compelling. Especially the kind of linguistic idealism in the later Wittgenstein of the Investigations. However there is a significant problem with Wittgenstein’s linguistic idealism, primarily that it can be unclear about the background mechanics of how the shaping of the world happens, that’s where grounding comes in.
Grounding is an explanatory dependence relation that seeks to solidify the relation where one thing is held “in virtue of” another thing. In the same areas where linguistic idealism can be too vague is grounding is explicitly clear. Attempting to deal with this relation in the everyday language environment where it comes up in the wild, means there are a lot of moving parts to grounding which I will discuss at length. There is also a fair amount of disagreement on some of those moving parts, so I will settle on Kit Fine’s treatment of grounding in particular, which seems to best allow for grounding to fit the everyday way that relation occurs.
I will be making some less popular decisions with regard to interpretation. Wittgenstein is generally not seens as a linguistic idealist. Some of why that is, has to do
with conflict with the Wittgenstein of the Tractatus, such as with Ryle and Findlay as I will discuss later. But Wittgenstein of the Investigations went to great lengths to distance himself from the Wittgenstein of the Tractatus, in particular referring to “the author of the Tractatus Logico-Philosophicus” (PI §23). And the parts of Wittgenstein in the Investigations that makes me inclined to agree with the minority saying he can be understood as a linguistic idealist are consistent through to On Certainty. Where he struggles with some of the weaknesses of linguistic idealism. I will discuss this consistency of the the late Wittgenstein more later.
The interpretation I will be using is that may strike some as strange is that I will often describe Kit Fine’s grounding as flexible. As I will discuss more later Fine has multiple ways he treats grounding, for example: weak grounding that contrasts against strict grounding, two approaches for different purposes. Fine has a preference for one but the other remains as a possibility. Fine is very specific about how each delineation of grounding works and within one distinction he will be strict about how far it reaches. But in having those different approaches I consider it flexible. Reflecting the differing but distinct ways we use language. Where some authors prefer to say there is one way to use grounding, such as Schaffer among others. But a restricted treatment of grounding results in being too strict and excluding a lot of what should be there, or trying to include too much under one kind of grounding together making it less effective. In allowing for the multiplicity of grounding methods Fine gets greater applicability without sacrificing scope.
Alongside the texts by Wittgenstein and Fine I’ll be focusing more on certain secondary sources than others. Anscombe, Bloor, and Dilman, primarily for Wittgenstein, Anscombe’s close relationship with Wittgenstein makes her a more trustworthy source for a controversial reading (Anscombe 1981: 125), and because she was among the first to present Wittgenstein as a linguistic idealist, most who follow her are building on this work of hers. Neither Wittgenstein nor Anscombe have a reputation of being excessively clear in their writing, I use Bloor’s paper revisiting Anscombe to help with that. And Dilman gives a clear and thorough look at the challenges to Wittgenstein’s linguistic idealism, and elsewhere often gets relied upon similarly to how I do. For Fine the main commentator I use is Fabrice and Schnieder, they give the most comprehensive and clear introduction to grounding that is consistent with Fine’s approach.
2.0 Wittgenstein's theory of meaning as use
First I’ll discuss Wittgenstein’s theory of meaning as use of the later Wittgenstein. How meaning is created plays a major role in how language functions for Wittgenstein, which is clearly important for how he can be read as a linguistic idealist. In this section I’ll also be establishing what I mean by “language”, as it has been treated in vastly different ways over the years and has a major impact on the plausibility of linguistic idealism.
Wittgenstein’s theory of meaning as use is most clearly demonstrated by his language games. He opens the Investigations with a primitive language game in §2 shaped to fit Augustine’s explanation of language getting meaning through reference. He describes in §1 a primitive language game between a builder and their assistant whose vocabulary limited to the objects they interact with. Wittgenstein brings in games to show how limiting of a definition of language this is, “It is as if someone were to say, ‘Playing a game consists in moving objects about on a surface according to certain rules.’ and we replied: You seem to be thinking of board-games, but they are not all the games there are” (PI §3). Just as there’s more games than board games there’s many more kinds of language than that limited to basic declarative statements. The issue at hand is how the words come to mean what they do, “How is what they signify supposed to come out other than in the kind of use they have? (PI §10). The way basic terms can be defined disguises how different their various usages can be, and so different meanings. Wittgenstein uses a metaphor of tools in §11. From just looking at a tool you might not know what they do, and even if given a basic description of what they do, you might not know how or when they’re being used correctly. This tool metaphor works especially well considering some basic tools you might easily be able to figure out like a hammer, but more complex tools like a lathe it would be quite the feat to figure out what it does independently without a demonstration. There are words like hammers and there are words like lathes, hammer words are the basic things that can be explained sufficiently to someone who doesn’t speak the language with a simple gesture, like “steep” referring to a hill. And there are words that are harder to explain, like technical or derogatory terms, that are like lathes. Even with a simple language, usage adds meaning to the words alone. As Wittgenstein raises in §19 the builder and their assistant after a while working together may develop a shorthand, so instead of the full order “Go get me a slab” he may just call out “slab” and the assistant would know what was meant (PI §20) . And as Wittgenstein remarks in §23 there’s many far more complicated ways we use language than the assertions Frege talks about. Wittgenstein includes examples like acting in a play as a use of language that wouldn’t fit Frege’s expectation of language being assertions with the assumption of truthfulness. Whereas when acting they’re intentionally being inaccurate, which is considered distinct from lying. This is why Wittgenstein calls them language games, there’s so many, and they’re often interrelated but not all directly, and frequently a new game will stem from an old one. Language games are generally1 developed through natural usage, in order for that usage to be successful there needs to be some commonality with conventional language use. One demonstration of this is in the etymology of words, several small iterations changing over time, language games develop in a similar way but on a larger scale.
Both individual and the general language usage are involved in the idea of meaning being developed though use. It’s a balance Wittgenstein draws with family
1
An exception to this would be constructed languages, like computer code, that didn’t gradually develop naturally but were explicitly set out to function in a specific way. These too would be language games despite their different development because of how they’re understood by their use.
resemblance. Family resemblance is a way of describing organization in a way that’s not as tight as categories tend to be, and not as loose as completely rejecting overarching terms. Both those approaches overlook how language really happens as a natural process of convention built by individuals. As Bloor describes it, family resemblance is Wittgenstein’s way of connecting to categories without sacrificing the flexible nature of language games. Given how language rarely has that kind of sharp boundaries around concepts and how they’re really used and how it changes (Bloor 2006: 354). This balance allows for terms to develop different meanings within a community, adding an extra meaning rather than replacing a general usage. For example how people can develop a shorthand on the basis of shared experience, the meaning for them doesn’t somehow get applied to when other people use the word because it’s not how they’re using it. Because usage is so active, the meaning in a language is constantly shifting, sometimes it’s just for a few people so it’s not a permanent change sometimes it’s big changes that last and add up to the little changes that brought Old English to the Modern English I’m using now.
There is a vast difference between how Wittgenstein understands meaning as the earlier Wittgenstein of the Tractatus, and as the later Wittgenstein of the Investigations. The difference is enough that when explicitly going against his earlier views he refers to “the author of Tractatus Logico-Philosophicus” (PI §23). I’m not going to get into exactly how Wittgenstein treated meaning in the Tractatus, it takes us too far afield of what I’m trying to say here. Suffice to say in the Tractatus is a realist approach and rather than including all the mess of everyday language in the Tractatus he restricted language to what can be contained within logic, isolated from the world it occurs in. Instead he treats language as an integral part of the world. A result of this view of language entrenched in the world is accepting the blurry nature of concepts, which he explores in respect to family resemblance. That blurriness is something Frege had a serious problem with, as Wittgenstein points out in §71 of the Investigations, Frege treated concepts as like regions which he insisted had to be well defined. But that insistence on clean borders just doesn’t reflect how people tend to think or talk. That discrepancy wouldn’t have been an issue for Frege or the Wittgenstein of the Tractatus, but by the Investigations Wittgenstein considers it a major problem.
How Frege’s discussions of identity statements in language remain in the realm of assertions seems to have lead Wittgenstein’s interest in the proposition. It pushed him to look at how propositions function as language in action. Wittgenstein departs from that view in the Investigations referring to Frege’s “every assertion contains an assumption” (PI §22). Wittgenstein argues that making explicit that a given sentence is an assertion as Frege suggests would be redundant, In the act of saying the sentence it becomes self-evident in the manner of saying an assertion that it is an assertion. Similarly with other types of speech like questions, what they are is clear in how they present themselves. The language is meaningful in it’s happening, the metalanguage comes packed in with the content. The assertion’s status as an assertion is made evident in its delivery, rather than having to be explicitly stated. This focus on the metalanguage that Wittgenstein inherited from Frege also shows up in Wittgenstein’s dismissal of ostensive definitions, which will
be discussed more later. In particular in §10 of the Investigations where Wittgenstein asks how it is specified what is signified by a given term when all you have is pointing, using an example of a blue vase and the ambiguity of pointing at the shape or the color.
Discussing Wittgenstein’s theory of meaning as use requires a lot of talking about language, and even more when discussing him as a linguistic idealist. The central claim is linguistic idealism is that our language shapes reality. What this means clearly changes drastically depending how language is understood, so first it’s important to establish what I mean by “language”. The way I use “language” is more inclusive than how some do, for me it includes more than the sounds and shapes of words spoken or written. This is consistent with how Wittgenstein refers to language in the Investigations, where the use of language forms an “important part of the life and culture in which the speakers of the language participate” (Dilman 2004: 164). This is a part of language that’s often not included but is vitally important when talking about Wittgenstein's linguistic idealism of the Investigations. As he put it in §23 of the Investigations before listing some of the ways language is used: “The word “language-game” is used here to emphasize the fact that the speaking of language is part of an activity, or of a form of life”. This is also where he distances himself from “the author or the Tractatus”. Wittgenstein understands meaning as determined by use, this automatically makes language an active thing. That use isn’t limited to the isolated sounds or marks that the words occur as, it’s everything that comes with the use. As opposed to language being separate from what is meaningful outside of the words as they occur in isolation. I think they’re too closely integrated for that, especially when discussing linguistic idealism.
Much of how we use words is context specific, in many senses of context. In particular as they arise in performative aspects of language, where the act of saying something or stylistic choices when writing conveys something. First what I will refer to as macro-context; the greater context the sentence is occuring in, where the larger society the utterance of the word or sentence is taking place in will have an effect on the meaning of a word. As an example, if I were to say “I’m going to check if that book is at the library” today I would be checking online, but around 20 years ago that wouldn’t be possible I would have to go to the actual building. Macro-context is also how references become integrated in the meaning, outside details involve themselves in how we understand a word or phrase. Macro-context also informs what Wittgenstein calls the “form of life”, where a person’s life shapes the language they use, which will be discussed more later (PI §23).
Language is also affected by the closer context within a sentence, which I will refer to as micro-context, how a word fits into a sentence will effect how it’s understood. This can be used to effect, like the classic Groucho Marx line “One morning I shot an elephant in my pajamas. How he got into my pajamas I'll never know”. The micro-context also includes the performative aspect of language which includes the actual speaking of words as well as what’s going on around that speaking, all the extra things that are why Anscombe prefers to use the more active sounding term and refer to “linguistic practices” over just language. This includes the auditory performance that involves things like
volume and tone, which as well as indicating the attitude of the speaker, tone also conveys meaning and is essential for things like sarcasm, all those things we struggle to convey in writing. There are also performative aspects in written language. Such as what gets referred to metaphorically as an author’s “voice”, markers in how the text is written so someone familiar with an author can recognize their writing, or an homage, without being told who it is. And there’s the rest of the performance of speech outside sounds that includes things like gestures, and the various behaviours around the speech or text that effect how it conveys. With linguistic idealism all those things contribute to how language creates the objects that fill our world.
2.1 Linguistic idealism
In this section I will cover what linguistic idealism is, in particular the linguistic idealism attributed to Wittgenstein, and how it’s appropriate to attribute it to him. The central claim of linguistic idealism is that our language creates the objects of the reality we live in. Wittgenstein’s language games and his views on meaning as use play an important role in his linguistic idealism. The active nature of Wittgenstein’s approach to language particularly lends itself to be understood in this way. Language is a living thing for Wittgenstein it does things, as I will discuss one of the things language does is create objects.
Different characterizations of linguistic idealism results in it swaying from sounding trivial to coming off as radical to the extent of being implausible, some of that comes from not being clear by what’s meant by “reality”. At the end of his paper “Wittgenstein and the Question of Linguistic Idealism” Dilman characterized linguistic idealism as making the claim of “a simple determination of reality by language” (Dilman 2004: 164). Reality here is not the physical stuff we interact with but rather the conceptual scaffolding around those things that inform how we interact with them. He’s not characterizing linguistic idealists as believing that if you call a tree a table and just keep saying it, the tree will transform into a table, bending to the demands of language. Rather the linguistic idealist is saying the reason we think of a particular piece of furniture one way, and treat it accordingly has to do with the language we use to describe it. Which is why Bloor characterizes linguistic idealism in a similar way: “Linguistic idealism is the claim that some truths or realities are created by our linguistic practices” (Bloor 2006: 356) adding that it can be understood that truths are what language is determining. Bloor there also uses the broader phrase “linguistic practices” that Anscombe also prefers over simply “language”. She also often uses the term essence where others might refer to reality (Anscombe 1981: 113).
One of the elements of linguistic idealism that can can lead to people dismissing it out of hand is “objects”, the things language creates. Objects have been talked about in many ways over the years, and I’m going to be using the term in a specific way. Importantly I don’t understand them as necessarily being physical. Sometimes objects will also be physical things like tables and trees and so on, in those cases that they’re physical is part of what it is to be that object. In other cases like unicorn, I still count them as objects, and physical presence isn’t part of what is is to be them. Things that don’t exist in all the many ways things can not exist, and still they’ll be meaningful. Not just fully fictional things like unicorns, but all the things referred to in counterfactuals, and things that don’t exist yet or anymore. This lack of commitment to the physical isn’t a problem for Wittgenstein’s linguistic idealism. If they are used successfully that’s all they need. They’re conceptually solid if not physically solid.
Dilman discusses how Wittgenstein has been understood as a linguistic idealist, and the different ways of accounting for Wittgenstein’s rejection that language is independent from the reality it expresses (Dilman 2004: 162). The various approaches Dilman discusses centre around different ways of recognizing Wittgenstein’s denial that the words of our language are based on some external reality to it. A common vein appears where several of the interpretations of Wittgenstein that Dilman dismisses are built around misunderstanding of the central role language plays, taking for granted a particular view of the way language is involved in the way objects are thought about.
Idealism in general, including Berkeley’s specifically, deals with the same problem as skepticism and realism: the question of how we get knowledge of physical things indirectly. While the skeptic and realist try to address it by rejecting either knowledge or the indirectness respectively, for the idealists Berkeley responded to this by reducing physical objects to ideas, going further than most people are comfortable with (Dilman 2004: 163). He calls into question the existence of a reality independent from our ideas of it. This can be taken to varying degrees, on the more extreme side rejecting the presence of the physical objects we seem to interact with, and on the other side claiming that the way we understand the world is due to the ideas we have informing the interpretation of those physical objects we interact with. The linguistic idealism applied to Wittgenstein here also needs to account for this, which it does in a different way, formulated as “There is no reality outside language and apart from its grammar or logical concepts” (Dilman 2004: 163).
Linguistic idealism is entirely in conflict with the linguistic realism which Dilman characterizes as claiming “that there must be simple objects, bare particulars, devoid of any property, which combined in various ways constitute what we state, describe, and speak about in the language we use” (Dilman 2004: 163). Dilman focuses on this particular version of realism because it’s the realism the Wittgenstein of the Tractatus holds, and because it has to deal with the issue he brought up of connecting meaning to objects. Wittgenstein in the Tractatus has his picture theory of meaning, where “the picture is a model of reality” (TLP 2.12) pictures are made of a collection of parts representing objects,
the structure of their combination represents a static state of affairs. These pictures are always a way of understanding the way things are in the world, in a somewhat limited sense. Truth then is determined by a better reflection. This covers a very limited section of all actual language use. The simple objects of realism act as the meanings of names, fixing their reference. This separates language and logic from their capitalized counterparts, Language and Logic supervening on them and existing apart from time and space (Dilman 2004: 163-164). In the Investigations this is completely revised, with language and Language reunited as it is firmly planted in language as it occurs in the wild. In the Investigations meaning is something created by usage, including all the messy ways humans use of language which is interwoven with the rest of the activity around it.
Dilman also looks at another interpretation ascribed to Wittgenstein, linguistic realism. It’s major departure from linguistic idealism is how the words are associated with their objects. Realism insists that the application of words to the things they name shouldn’t be arbitrary. There should be a common thing being named that is why all the same things get called the same word: “One does not use the word any old how. There must be something about those instances, something that is true of them all, which makes the word applicable to them” (Dilman 2004: 164). This attitude towards language is common in philosophy, to look at the language as it is, ignoring how it got that way. How arbitrary language can be is best demonstrated in slang, which appears and changes and disappears so quickly it can be difficult to keep up and yet it is understandable to those using it. Also how Old English isn’t readable to modern English speakers, and yet it became the language we have today over a series of small changes over many years. Linguistic realism requires something separate from language for language to be applied to. Dilman refers to Price, who in Thinking and Experience argues repetition in the world is how we develop conceptual cognition (Dilman 2004: 164). Though Dilman points out this overlooks the obvious phenomena that is prevalent even among people totally fluent in a language, namely that we can live our whole lives around things and not know the names for them. For example trees, a lot of people don’t bother to learn about specific kinds of trees and it’s very easy to go through your life without it mattering. Repeated exposure to a tree doesn’t lead to any kind of enlightenment to what it’s called, there’s nothing in the tree itself that indicates what it’s called. This is the problem commonly known as “the one and the many” how general terms develop their meaning through individuals, and is of great interest to nominalists. So the approach to language in linguistic realism isn’t particularly consistent with Wittgenstein and how he treats language, linguistic idealism works much better.
Taking a different approach, Anscombe frames her argument that Wittgenstein is a linguistic idealist around the treatment of essence by Wittgenstein embodied most clearly with “essence is expressed by grammar” (PI §371) as well as in Wittgenstein’s Remarks on the Foundations of Mathematics “it is not the property of an object that is ever 'essential', but rather the mark of a concept” (RFM I- 73). So the essential part of concepts, the meaningful part, is embodied in linguistic practices. As opposed to the physical things being where the essence of these concepts are. So that without the word “horse” the actual
animals would still exist and be perfectly happy, but the concept of “horse” that makes them horses would not be out there somehow. She treats the idea of essence in grammar as a natural progression of Wittgenstein’s participatory language that has always been present in his writing about language.
How Wittgenstein understands meaning is tied to how he can be seen as a linguistic idealist. For Wittgenstein how we establish meaning is how we understand the essence of what we’re talking about. It comes out in the often referred to “essence is expressed by grammar” (PI §371). Grammar for Wittgenstein isn’t prescriptive, it acts as the established patterns in language, built through usage constantly changing: “It only describes, and in no way explains, the use of signs” (PI §496). Linguistic idealism hinges on the relationship between language and the things of the world, as well as how we understand both those parts. When Wittgenstein is read this way his views of how the meaning of language is created, in a way are also views of how the world we live in itself is created. The step between essence being expressed by grammar and the creation of objects by language is in the role grammar plays in language being successful. Grammar is the convention of how we organize language, establishing some consistency within a language game. That consistency is needed to get enough consensus for an essence to be established, "The essence is not what I mean or am speaking of: it is rather that through which I understand or whink of (mean) etc" (Anscombe 1981: 115). Grammar plays a role in establishing the meaning of things by regularizing the context the words show up in, which is particularly important in solidifying homonyms in written language like the past and present tense of “read”.
Language for Wittgenstein is a thing created by doing it, by interacting and participating with it. And grammar is the way that is developed to indicate what kind of thing we’re dealing with, a structure that allows for more clear communication; a constructed microcontext to indicate usage. The grammar is developed through the successful usage of a language. Because a language isn’t handed down as a complete whole, instead we add words to our language when we come across them. And often we don’t require a definition for a new word, how it fits in the rest of the sentence provides some information of the kind of thing it refers to. For example if someone were to say to you “yesterday I got a beech sapling” even if you didn’t know about beech trees you probably wouldn’t get “beech” mixed up with the homophone “beach”. As he puts it when discussing rules in games “a rule is employed neither in the teaching nor in the game itself; nor is it set down in a list of rules. One learns the game by watching how others play it (PI §54). In grammar also, rules are not always explicitly drawn out. An important aspect of how this works is raised in the box remark after §138 asking when we know a word. As with many things a lot of light can be cast on something by asking what’s happening when something goes wrong. Wittgenstein does this when he raises the problem of misunderstanding, when we think we know what a word means but are wrong. Here, where Wittgenstein is distancing himself from the picture model of language he has in the Tractatus, is one of the places Wittgenstein most clearly connects with the linguistic idealism Anscombe discusses. When usage is how we develop the meaning of concepts we
run into an issue because humans aren’t computers, we’re not always perfectly consistent or clear, and often leave bits of information out for the sake of saving time. This is why we will need to encounter a word a few times before really feeling like we know it. The process of language determining the objects of the world requires first for the development of meaning.
Clearly usage is human dependent, and so creating concepts through developing meaning addresses a lot problems that are present in other ways of explaining meaning such as his previous referential picture theory of the Tractatus. Which Wittgenstein addresses when he discusses of the problems with defining words ostensibly, most importantly is an issue with authority to consensus, and size of available vocabulary. This isn’t a problem when words are defined through usage. When in §373 of the Investigations Wittgenstein states “Grammar tells what kind of object anything is.” it connects with §73 of his Remarks on the Foundations of Mathematics where by way of “Red is” as a proposition Wittgenstein discusses how convention is deeply integrated into the essence. For Wittgenstein it’s not just “merely” linguistic convention, it isn’t something that can be waved aside. Convention is how concepts are created. Ultimately for Wittgenstein language is built on how we use it. And in turn that language determines how we understand the world, in how we decide how to break the world up into kinds or keep sames together (PI §378). The exact mechanics of how that creation works is less clear, and that is something I’ll be addressing later.
Wittgenstein says “the speaking of language is part of an activity, or of a form of life” (PI §23) but he isn’t particularly clear about what he means by “form of life”. It seems that by form of life he means that participating in a particular life you lead shapes the language games you participate in. And the experience of your life makes certain parts of the world accessible to you: “to imagine a language means to imagine a form of life” (PI §19). The form of life also comes up when confronting the issue of arbitrariness of truth when meaning is determined by usage, an issue I will discuss more later. To which Wittgenstein responds “What is true or false is what human beings say; and it is in their language that human beings agree. This is agreement not in opinions, but rather in form of life” (PI §241). This is a major point where Wittgenstein connects to linguistic idealism. He seems to be saying here that those agreements between language and life don’t just correlate with reality, but in agreeing together determine something about reality. He goes on to discuss consistency in measurement, so I don’t think he means that in as radical of a way as it may sound. I don’t think he’s saying that if enough people agree that a tree is actually a porcupine it will become so. Rather that all the things we think of when we think of trees, forming our idea of what it means to be a tree are created through our use of the word “tree” and the language we surround it with.
Something significant about essence for Wittgenstein can be demonstrated with indexicals, because they put up the most resistance to being defined as signifying through reference. In §46 Wittgenstein brings up Plato’s Theaetetus representing a contrasting view on indexicals, where they call words like “this” a primary element, as it’s simple. “This” is
not connected to a concrete thing the way a word like “table” can be argued to be connected to physical tables. Wittgenstein relates this to Russell’s individuals as well as his objects in the Tractatus (PI §46). But he quickly dismisses this idea of primary elements in language as the most essential part of the thing being understood (PI §47). In §47 Wittgenstein asks what level of a chair is the most primary, it’s clearly made up of things, but by seeking where it isn’t composite one can easily slip down to thinking of the chair’s particles are where it’s essence lies. But in looking at the particles that make up the wood that makes up the chair there’s nothing particularly chair-like in them. That those particles of wood are part of a chair is incidental. If that piece of wood had been used to make a table nobody could see it and say there’s a bit of a chair in this table. The chair’s chair-ness is in being a complete chair that is talked about as a particular type of thing. The way we talk about the chair, is what makes what we think of as a chair, be a chair. The chairness of a chair depending on how we talk about it is particularly well demonstrated with abnormal instances, when a chair that doesn’t necessarily look like a lot of other chairs is a chair because of how it’s being treated and talked about. This is what Wittgenstein means when later in §371 he says “essence is expressed in grammar”.
Empiricism also falls into the problem of assuming it’s perceptual data to be prior to the language it uses to describe those perceptions. And uses that priority as a measure by which to compare sense based statements. In order for those statements about sense phenomena to be meaningful they need to have language attached to it, so that comparisons, hypotheses, and conclusions can be drawn. When Wittgenstein refers to the limits of empiricism it’s focused on this element, the empiricism of physics and other common uses of empiricism, more than philosophical empiricism. By attributing truth to a proposition goes beyond its scope. The problem is in “confusing what makes a proposition true with the reality we take for granted in referring to what makes it true” (Dilman 2004: 171). What the truth really depends on is the language. What it expresses comes in through the language not the thing being referred to. Again when looking for ultimate justification they look in the wrong place. The empiricist “mislocates the concept of physical reality -a formal concept- by treating it as an ordinary, empirical concept, signifying a class of objects such as trees, houses, etc”, but those designations of sets of trees and houses and so on are established by language (Dilman 2004: 172). As is the chain of classification we tend to use with nesting sets: “a whale is a mammal, a mammal is an animal, an animal is a living thing, a living thing is an organism” (Dilman 2004: 172) and so on. It’s easy to forget these classifications are a product of language, because they become part of the concept involved with what they’re classifying. And so it’s easy to forget there was a time when the idea of mammal didn’t exist, and it was much more recently than when the beings themselves didn’t exist. Empirical judgements depend on the language they take place in, it determines the kinds of observations that can be made, and the conclusions that can be drawn from them. “It is not what we experience that determines the kind of language we speak; but rather, the kind of language we have developed, or better the kind of language that has developed in a society or community determines the kind of experience that we have, as members of that community (Dilman 2004: 172). Empiricists often view themselves as entirely independent of outside societal forces leading them
astray with subjectivity they think their sense data is free of, but they are still language users. As such they still understand their findings from within language, interpreting phenomena with terminology, which forms the reality they’re trying to get past. So by attempting to bypass language they’re ignoring what makes what they’re looking at a something: “Both the uses of the senses and the use of language are part of our natural history and they are intertwined” (Dilman 2004: 172). Separating the senses from the world of language and its history isn’t going to be fruitful, instead as Kant says they end up blind.
Religious use of language under linguistic idealism takes an interesting flavour, Dilman refers to Winch with “But it is within the religious use of language that the conception of God’s reality has its place” (Dilman 2004: 173), it’s created in the language about it. Leading to the question of how different the realities of different people will be when they can have such big differences in how, or if, they use a kind of language. Wittgenstein’s understanding of language as use isn’t entirely individualistic, but it does also depend on the individual to contribute to the collective. However how much of a collective is needed, and to what extent is the internal experience of the individuals the same or different from that of others isn’t particularly clear. The kind of life the individual leads will have a big impact, “Many of the things with which we engage and that are, therefore, real to us, have no reality for the cat. They do not exist in or form part of its world” (Dilman 2004: 174). For example: taxes, a cat has no experience paying taxes and has no idea they exist for us, they’re not part of the cat’s life and so don’t exist for the cat. Linguistic idealism makes the claim of “a simple determinism of reality by language. What Wittgenstein brings out is that while our conception of reality is internal to our language, that language itself is rooted in a life which the speakers share” (Dilman 2004: 177). That reality is informed by the concepts any individual comes in contact with and becomes part of their vocabulary. The results are going to be different for different groups of people but the mechanism of how a person’s reality is formed remains the same, and it informs what Wittgenstein would call their form of life. The impact the difference between people’s lives comes up notably references in poetic language, where references will often be subtle and easy to miss if you’re not familiar with what’s being referenced without even realizing you’re missing something, and that changes what the words mean for the reader.
Anscombe provides rules as an example of something that clearly relies on linguistic practice, and so act as an uncontroversial example of how linguistic practices have clear effects. Rules are a mechanism of specifying behaviours as allowed or not. She distinguishes rules from promises, which can be described in a very similar way, by how promises are explicitly stated and agreed upon (Anscombe 1981: 118). Whereas rules tend to be already established and aren’t necessarily stated outright. Yet there are expectations around rules reflected in the linguistic practices surrounding them. So rules are something Wittgenstein spends a great deal of time on, as they’re involved in language games2. Wittgenstein finds the common assumption that rules are meant to act as means of
2
interpreting language with the purpose of learning to predict it problematic. His problems with that view are brought up in the paradox in §201 where anything can be said to be according to a rule because a rule can always be formulated to include whatever may be the case. This is why he treats rules of this kind as a means of interpretation, they are a way we understand these linguistic practices. And they are intrinsically performative which is why Wittgenstein says that rules can’t be followed in private. As essential part of the rule is in the interaction with others: “That’s why ‘following a rule’ is a practice. And to think one is following a rule is not to follow a rule. And that’s why it’s not possible, that’s why ‘following a rule’ is a practice. And to think one is following a rule is not to follow a rule. And that’s why it’s not possible” (PI §202). Language games for Wittgenstein involve rules that describe separate instances of language that belong in a language game, but together what makes something a language game resists sharp boundaries. It’s a family resemblance, much like many linguistic practices for Anscombe. The core of what makes a practice or a game work might be easily enough defined loosely but to exhaustively define it is another matter: “Where the regular manner of acting defines a practice, we may avoid the problem by saying that the rules of course don’t impose any necessity upon you, except that of acting so if you wish to engage in the practice” (Anscombe 1981: 120). So a rule of language applies when you are intending to follow it. These rules help develop what in On Certainty he will call knowledge systems, as he seeks something under these rules, which I will discuss later.
2.2 Some criticism with responses
Reading Wittgenstein as a linguistic idealist is a controversial stance, so naturally there are criticisms. Some of those criticisms I will addressed here, those that largely have to do with what one understands “language” to be. And some about the ambiguity around the mechanics of linguistic idealism actually doing what it claims to do, I’ll address later.
Alongside Ryle, Findlay criticises the view of meaning as usage through the old saying Ryle quotes as: “Don’t ask for the meaning, ask for the use” (Ryle, Findlay 1961: 229). He ignores how in the saying “meaning” is meant more like “definition”, but both Ryle and Findlay treat it in the more broad sense we normally think of meaning. Instead Findlay follows Ryle’s understanding of language and meaning which Findlay describes as such: “I think we want to distinguish between [language,] a sentence as a grammatically permissible word-combination, and the utterance or writing down or silent thinking of that sentence by someone on some occasion to make an allegation [...] the latter case I find a language of use" (Ryle, Findlay 1961: 231). Findlay claims that usage “presupposes the notion of meaning (in its central and paradigmatic sense) and that it cannot therefore be used to elucidate the latter, and much less to replace or to do duty for it (Ryle, Findlay 1961: 233). He goes on to say that usage covers more than meaning, so meaning can’t be reduced to usage. Findlay also goes on to discuss a difficulty with saying how a word is
used, claiming a difficulty in including things like what it’s meant to refer to. He seems to have an impression that giving the meaning of a word with its usage would take a similar form as a dictionary definition, rather than examples of the word in a sentence. Wittgenstein isn’t claiming that dictionary definitions should be entirely replaced by just an example of the word being used, but rather that those definitions come after the meaning is established through usage. By determining meaning by use Wittgenstein isn’t reducing meaning to be solely usage. Instead he’s expanding meaning by including usage, and recognizing language as a living thing.
Dilman discusses how Brambrought tries to avoid calling Wittgenstein a nominalist by trying to interpret him as a realist, mostly drawing from §66 of the Investigations where Wittgenstein mentions similarities in the things we call games. Though in the immediately following section Wittgenstein calls that similarity family resemblance leading to his language games as he rejects his earlier ideas of the form of a proposition from the Tractatus. The idea of a common essence that connects words to their things can seem accurate to what’s going on, but we have to remember that we’re looking at this as language users. It’s because of how general terms work that we want to think they’re all referring to something common separate from any instance (Dilman 2004: 165). But as with Wittgenstein’s example with looking for the real artichoke beyond its leaves (PI §164), when we try to find the real meaning beyond the instances of a general terms we’ve gone past where the real meaning is; the leaves are a vital part of the artichoke, it can’t be found without them. But of course Wittgenstein isn’t being so straightforward to just say that meaning of general terms are entirely to be understood by their individuals. As he later shows when talking about the smile, (PI §583) and the difference between the various ways a smile can be understood as friendly or not, which is determined by the situation outside of the smile itself.
Bambrough in recognising Wittgenstein isn’t a nominalist struggles to see how he doesn’t leave language “hanging in midair” (Dilman 2004: 166), the answer to which is that Wittgenstein teathers language to usage. Both in terms of instances and in the generality of agreement of how those instances should be used, something that fluctuates and adjusts over time. Nominalism and realism are both attempts to find “ultimate justification” of words that can be an indicator for how language functions.
In both Bambrough and Findlay I think there is a significant misunderstanding of what Wittgenstein is saying when he defines meaning as usage. I will be discussing their complaints against Wittgenstein linguistic idealism first, then responding to them later.
The main confusion centering around how Wittgenstein’s not reducing usage to meaning, he’s expanding what is meant by “meaning” to include usage. Wittgenstein would agree that usage contains more than a definition of the word. That’s what he is trying to do with language games, include all the parts of how words are understood in the word’s meaning. Bramborough seems to think usage is far more ephemeral than Wittgenstein means it, but it’s tethered in consensus. I think some of this misunderstanding
is coming from Findlay treating Wittgenstein’s view of meaning in the Investigations and the Tractatus to be consistent. Findlay attempts to connect the stance of the Investigations where words having meaning through our use of them with “since understanding and thinking are defined in terms of the operation with signs, we must never speak as if we could understand or think anything before we dispose of appropriate verbal expressions and have been taught to employ them” (Ryle, Findlay 1961: 236). But that is more in line with the Tractatus, when Wittgenstein’s departed so far from his views in the Tractatus that in the Investigations he occasionally refers to the author of the Tractatus as if that’s not him.
Essence plays a very significant role in linguistic idealism, it is what is being created by these linguistic practices. But it’s not particularly clear exactly what that is, which contributes to how unclear how significant the claim of someone’s linguistic idealism really is. There seems to be a very fine line between the trivial linguistic idealism that claims the concepts are what’s created by language, and the extreme side of what can be meant by linguistic idealism, claiming that language is responsible for the existence of physical objects. There’s such a vast difference between those two claims it should be very clear which is being meant when, but it’s often not particularly clear at all. I think a lot of that has to do with not making explicit the role essence, or something similar, plays in it. And since essence isn’t an uncontroversial idea, what the particular author means by essence. Wittgenstein isn’t entirely clear what he means by essence, and seems inconsistent about it at times. The element of essence that he is consistent about is the role it plays with grammar. In §89-92 Wittgenstein describes essence as foundational, and concerned with how things are what they are, this leads towards “the ‘possibilities’ of phenomena” rather than the phenomena itself. Possibility is where misunderstanding can come up, which he addresses with grammar. Possibility is like analogies providing synonymous description, and lending itself to being pulled apart and analysed. Essence is what the grammar reveals when pulled apart, this is finally put succinctly later by §371 “Essence is expressed by grammar”.
What Wittgenstein is less consistent about with regard to essence, is its availability. In §92 he puts it quite plainly “For [the question of the essence of language] sees the essence of things not as something that already lies open to view, and that becomes surveyable through a process of ordering, but as some-thing that lies beneath the surface”. This follows Wittgenstein’s reference to Augustine in §89, where someone feels like they understand something when it’s used but when pressed for a definition isn’t confident giving one. The essence as something we grasp, and is there in our ability to give analogies and synonyms to clear up misunderstanding, but can be elusive if we try to look right at it itself. But then later in section 116 the essence seems to be something in the everyday, warning against over-conceptualizing words like knowledge and being Wittgenstein asks: “is the word ever actually used in this way in the language in which it is at home? What we do is to bring words back from their metaphysical to their everyday use.” (PI § 116) when he goes on to discuss how the meaning in language is found in the everyday use, rather than prescriptive applying definitions. Which since he earlier treated essence as being a
core part of meaning, in the possibilities of phenomena and what is in common in synonyms it should be something available. Wittgenstein seems to recognise this in §92 he frames the problem to address as “the essence is hidden from us”, which leads into a discussion of propositions, largely involving the distinction of sign and fact in an account of logic. And he struggles again, in a similar way with what is out front and active, and what is behind it.
A lot of that confusion with the accessibility of essence is cleared up within different understandings of Linguistic idealism. Generally leaning towards essence being more accessible because of how integrated it is with the language. Not just because of how that language is something we interact with everyday but because that language is how we create our reality. What it means to create our reality is where linguistic idealists disagree. But if language is to play an active role in the way we understand our world then that requires some access to the mechanism by which that happens. And that mechanism is essence. As Bloor puts it “linguistic idealism implies that some “essences” can be created in the course of being expressed” (Bloor 2006: 356). While this can be seen as somewhat circular if they’re not careful the meaning in language will seem to both create essences and be informed by essences. With Wittgenstein it’s not necessarily a problem, when meaning is use this idea of creation when expressed is more of a feature than a problem. Exactly how that creation when expressed works when you try to look directly at that step isn’t particularly clear, and it’s what I hope to add clarity to later on.
There are other versions of linguistic idealism that don’t fit with Wittgenstein of the Investigations, such as Gaskin’s which “holds that the world consists of true and false Russellian propositions, presented by true and false declarative sentences mediated by true and false Fregean Thoughts” (Gaskin 2016: 1). This limitation to declarative statements and a more hard and singular approach to truth in meaning is drastically different from the linguistic idealism of Wittgenstein. While Gaskin’s focus on the proposition looks similar to Wittgenstein, what’s included in that is drastically different. As one case of disagreement the example Wittgenstein has of different kinds of smiles conveying different meanings (PI §583), wouldn’t be included in Gaskin’s idea of language. As Linsky points out in his criticism of Gaskin much of his unity of the proposition comes from the syntax (Linsky 2011: 469). Something that is not present in Wittgenstein’s Linguistic Idealism.
2.3 Some advantages to linguistic idealism
Amid the many criticisms it’s important to highlight why we would bother to try to fix those problems. In this section I’ll be discussing linguistic idealism’s strengths, mainly how comfortable linguistic idealism is in the wild of everyday language; how flexible is with the many ways language is used.
Dilman connects the issue of identification of a kind of thing and classification, with Wittgenstein’s argument against the ostensive definitions in the first several sections of the Investigations. The description of ostensive definitions Wittgenstein gives comes from Augustine (PI §1) as an explanation of language development. Demonstrated with a story of learning words from some authority who establishes a connection between a word and an object that it refers to by pointing and naming. As Dilman points out this requires the prior existence of language for the words to be there to be drawn from and applied to a thing. There has to be a name and an authority to be in the position to definitively say it is the word for the thing, and how it fits into the greater language. Another issue with Augustine’s view of language is that classification requires objects: “we do not, as it were, classify bare particulars, we do not state in a grammatical vacuum. And where we have objects to classify, we have already names for them (Dilman 2004: 167). We have to first know something about a thing in order to recognize how a thing fits in the general scheme of how we organize all the other things we know about. Dilman gives the example of knowing what ‘elm trees’ are requires knowing how they are distinct from ‘trees’ generally. Of course a specific definition of things can take time to establish, this is easily forgotten as much of the things we like to use for example objects like trees and tables are part of our common vocabulary with a lot of instances. When it’s something we deal with less often and have fewer instances it gets murky. This was the reasoning for Pluto being demoted from planet status to be a dwarf planet in 2006, our definition of planet wasn’t very specific and with so few cases to differentiate things that are close can get confused.
Ostensive definitions also fall into the problem of nonbeing. How to point to and name something that doesn’t exist is difficult under such a limited explanation of language acquisition that relies on direct reference. Successfully referring to unicorns would be impossible, and yet we manage it quite easily, I can use the word and be reasonably sure you’ll know what I mean. And it’s not just a very charitable approach to admitting things meaning. If when trying to refer to a unicorn I say “A unicorn is a large flying reptile that can breathe fire” you’d be likely to disagree, and rightfully so as that’s a description of a dragon, a completely different nonexistent thing. This isn’t a problem for linguistic idealism, because the objects in question, existent or not, are dependent on the language. For ostensive definitions it’s the other way around, the words need something present to attach themselves to. Nominalism as a form of linguistic idealism as Dilman describes it as doesn’t avoid the problem that Bambrough described as letting language hang in midair, by refusing to admit universals, and so also essence. Because some words can’t be found in concrete space nominalists are left without a way for language to be connected to what it’s about. But this isn’t a problem for Wittgenstein’s linguistic idealism, as essence has an important part in how all this works for him. For Wittgenstein “reality is a relative term, it is internally related to language, internal to the grammar of paricial modes of discourse which form part of a natural language. But a natural language is rooted in the life of the people who speak it” (Dilman 2004: 169). For Wittgenstein language and reality are intrinsically connected, as language is the way reality comes to be meaningful, and not just in a sterilized formal version of language but the actual language used every day by people as part of living their lives. A big part of how this plays out is how “essence is expressed
by grammar” (PI §371) that process of language, in all it’s capacities housing reality, is how essence connects to language through grammar in the flexible structures of language games. Anscombe suggests understanding that connection between essence and grammar in a way similar to Plato’s Cratylus. Where grammar is like “a tool which is designed to catch hold of something will perhaps have a shape corresponding to the shape of the object” (Anscombe 1981: 112). This relationship between grammar and essence for Wittgenstein is one that is constantly growing as language shifts and the way we use words change, creating new tools and changing old ones. In the action of life essence comes to light.
3.0 Introduction to grounding
While grounding has only really come into its own in the last decade or so it has roots that go beyond that, first I will discuss those roots before moving to the contemporary grounding discussion. Starting with Plato’s Euthyphro, and Leibniz’s Principle of Sufficient Reason, and Bolzano’s view of grounding as an entailment relation (Correia and Schnieder 2012: 2).
Euthyphro is where Socrates asks if something is pious because it is favoured by the gods or if it’s favoured by the gods because it’s pious. Socrates doing what he does best leads Euthyphro to contradict himself, by trying to get at how piety is grounded, what a pious act’s piety is in virtue of. In that argument Socrates raises many of the major principles of grounding as we have it today: “It involves one substantive grounding claim (relational properties are grounded in relations) and it implicitly draws on structural properties of grounding, namely symmetry and transitivity” (Correia and Schnieder 2012: 3-4).
Leibniz’s Principle of Sufficient Reason put simply, claims everything needs to have a sufficient reason to exist or happen. He examines how and when the connection of dependency is made. What is meant by a reason is a matter of debate among rationalists (Correia and Schnieder 2012: 4). Mainly between whether they are reasons for the things in question, or reasons for knowing a fact about the thing (Correia and Schnieder 2012: 4-5). The option of the sufficient reasons in question being about the things fits more smoothly with grounding as it’s understood now. Because the Principle of Sufficient Reason doesn’t allow anything to stand on its own without grounds, it denies fundamental facts. That rejection of the fundamental was popular at the time it is now not uncontroversial, and grounding doesn’t hold onto that part of it. But the idea of sufficient reason of reasoned dependence is still at the core of grounding.
Correia and Schnieder also discuss Bolzano’s understanding of grounding as an entailment relation, “propositions as abstract objects which are structured compounds of
concepts and potential contents of judgements and assertions.” (Correia and Schnieder 2012: 6). That structure is grounding, and it creates an explanatory force in this relation. That force can be applied regardless of subject matter; mathematical as well as philosophical. Bolzano’s idea of grounding also came up with his view on a Supreme Moral Law, where he describes foundational truths: “It must therefore be a basic truth, i.e., a truth that does not have an objective “ground”, but can only be an objective “ground” of other (practical) truths.” (Morscher 2014: 5.2) However Bolzano mainly focused on logical grounding, being focused on specifically “objective ground” he is concerned with logical entailment rather than a metaphysical or epistemological one (Morscher 2014: 4.5). The distinction between logical and metaphysical grounding is largely based on the role of truth. For Bolzano’s logical grounding it starts with a requirement of truth, and then examines how one got there. As such it is about how truths are arrived at, rather than how propositions rely on each other. Bolzano considered certain mathematical proofs one way of uncovering grounds “if they not only demonstrate that a certain mathematical truth holds but also disclose why it holds” (Correia and Schnieder 2012: 6-7). Carrying across much of the attitude in logical grounding Bolzano included in his metaphysical grounding a principle of “Factivity [which he defines as] grounding connects only true propositions” 3 in the “crucial structural features” of grounding (Correia and Schnieder 2012: 8). Whereas Fine’s metaphysical grounding is about the propositions just as propositions and how they connect, rather than being too concerned with the propositions being true. Fine’s approach opens it up to being applied more broadly, as Fine will apply grounding to mathematics, and ethics, and so on. Grounding understood this way can therefore be used to examine beliefs that depend on each other, without requiring that every one of a person’s beliefs be true, or consider counterfactual scenarios. The role of truth largely depends on their stance on what the relata of a grounding relation is, which I will discuss more later.
Moving on to the contemporary discussions about grounding. In the intervening time the notion of a proposition has become a much more significant focus in metaphysics, and accordingly has taking up a larger role in grounding.
Fine describes grounding as explanation: “we take ground to be an explanatory relation: if the truth that P is grounded in other truths, then they account for its truth; P’s being the case holds in virtue of other truths’ being the case” (Fine 2001:15). It is not the only explanatory relation, but Fine calls it the “tightest such connection” (Fine 2001: 15). The explanatory connection between the two propositions is really what the grounding relation consists in: “The grounds must account for what is grounded. Thus even though P is a consequence of [Q&R], [Q&R] is not a ground for P, since it does not account for the truth of P” (Fine 2010: 1) . To really be in virtue of something the connection to that thing needs to be stronger, the ingredients of a grounding relation can’t be arbitrary, it has to be solidified with the help of something. And that something is not causal, which I’ll be discussing more later. Grounding functions as a “sentential operator, in the same way as ‘if
3
Factivity as Bolzano includes it is distinct from factualism which I will discuss more later. The main difference is in the role of truth, which is central to factivity isn’t necessarily as big of a concern to factualism.
then’” (Fine 2001: 16), and so the relation of grounding exists on the level of the proposition, and so grounding “need not be seen as engaging either with the concept of truth or with the ontology of facts” (Fine 2001: 16). Fine wants his grounding to work with propositions as propositions, just the units of language they are. He doesn’t want to treat propositions as vehicles for Truth, or propositions as references to the state of the physical world, or anything else: “insofar as [ground] is regarded as a relation, it should be seen to hold between entities of the same type" (Fine 2012a: 43).
Gounding also requires a stance on factualism, to decide how questions of ground would be settled, and whose “account of practice” (Fine 2001: 16) works better. While factualism is central to the mind independence of realism, factualism isn’t inherently realist itself. Factualism requires that propositions express some (Khlentzos 2016: 1). What one considers to be a fact will greatly impact how that general stance really takes effect when connected to something like grounding. Namely whether those facts require being connected to the world in some way, or if facts are just something that has a truth about them. Factualism’s reverse, anti-factualism can also be referred to emotivism, which claims that propositions don’t reflect some fact but instead are really just statements of feeling towards the subject (Joyce 2016: 3). Different versions of grounding connect to factualism or anti-factualism by deciding what is the relata; the kinds of things that are eligible to ground other things. An anti-factualist stance is in conflict with the very large view of language discussed earlier, including macro and micro context along with words, by reducing all the detail in the meaning of something to just a basic emotional state loses so much nuance. Fine specifically points out “nondeclarative pronouncements —such as “Ouch!”,” (Fine 2001: 4) as an example of language use that while not being a clear statement of fact has a fact of the matter within its meaning. This flexible treatment of language that includes things without obvious factual status fits with grounding. That grounding is flexible is helpful as it’s a relation that gets used in a lot of ways.4 Grounding tends to be factualist, as at its core it tends to be about the interaction of propositions.
This factualist understanding of grounding may sound like it’s at odds with the way I referred to how Bolzano’s Factivity relates to Fine’s grounding earlier but it’s not. Factivity is about there being a requirement for truth at the outset. Whereas the facts referred to be factualism aren’t necessarily what is being said. The meaning of a statement being different from what is uttered is a regular occurrence out in the wilds of everyday language. Discrepancies in meaning and spoken words are seen in nondeclarative use of language, such Fine’s example of “ouch”, or a gesture, or Wittgenstein’s example in §581 of the versatility of a smile. There is a fact corresponding to a false statement it just isn’t what is said. For example a statement like “there’s are 23 planets in our solar system” has a
4 Correia and Schnieder demonstrate grounding variations with a list of several ways that things are grounded, this is reminiscent of §23 of the Investigations where Wittgenstein gives examples of ways language is used. The examples of usages of grounding such as “A set of things is less fundamental than its members” (Correia and Schnieder 2012: 1) echos examples of language games like “Describing an object by its appearance, or by its measurements” (PI §23). Both language games and grounding are concepts that are most succinctly explained through examples, they’re both deeply integrated in our experience that pointing out that is more effective than a definition.
fact connected to it, namely the actual number of planets. But a different wrong number of planets like 9 also may have a historical fact connected to it of when that was the commonly held belief of the right number of planets. Factivity however would make false statements not eligible to be part of a grounding relation, which limits how grounding can be used.
Grounding is a relation that is often taken for granted in philosophy, and as Rosen points out grounding is sometimes dismissed as just an idiom of our language, but really it needs to be explored (Rosen 2010: 109). Rosen discusses how some dismiss grounding by claiming it’s really just idiomatic, not a legitimate philosophical relation, admitting there’s no reduction to be done there “We do not know how to say in any more basic terms what it is for one fact to obtain in virtue of another” (Rosen 2010: 113). But that seems more an argument that the relation is foundational not that there’s nothing to examine there. Being a primitive relation doesn’t make it “unclear or unintelligible”, as it can be accused of being. Rosen goes on to make the point that a value of grounding can be using moral facts supervening on social facts being something both a positivist and antipositivist will agree on, but they will disagree on whether moral facts play a role in making laws (Rosen 2010: 113-114). What is being disagreed on there is how the making of laws is grounded in moral facts or not, in how various facts interact and the roles they play in each other.
So far I’ve focused on what grounding does before shifting to focus on a specific kind of grounding I’ll take a look at how it actually accomplishes that. Being a relation there are distinct rules of how it interacts with its relata. There’s still a lot of disagreement about these but as the way he understands grounding works better for what I’m going to try to do with it I’ll focus mostly on Fine’s approach. A major feature of Fine’s approach to grounding is its flexibility that isn’t just vagueness, but well defined through a hierarchy of grounding distinctions: full/partial, strict/weak, and immediate/ mediate which are interconnected and I will discuss in turn (Fine 2012: 51).
Transitivity is generally recognized as holding through grounding, which is part of why chains of grounded and grounding frequently come up. For Fine it’s an element of his weak partial grounding. Weak and partial grounding both look at how propositions are often held in virtue of multiple propositions, together covering both sides. Weak5 grounding allows for a proposition to be held in virtue of any number of propositions: “the grounding operator "<" is variably polyadic; although it must take exactly one argument to its "right" it may take any number of arguments to its left" (Fine 2012a: 47). Connected to weak ground is partial6 grounding which allows for a proposition to only account for some of the grounding of another proposition (Fine 2012a: 50). Fine also considers an intransitive version of grounding which he calls immediate ground (Clark and Liggins 2012: 814), contrasted with mediate ground. Immediate ground doesn’t allow chains of
5 The converse to weak ground, strict ground requires that a proposition can only be held in virtue of one other proposition. (Fine 2012a: 51)
6
The converse to partial ground, full ground requires that a proposition be held entirely in virtue of a single proposition. (Fine 2012a: 50)
