Tuesday, March 18, 2014

Language as a Tool for Contemplation and Communication

I. Language Development and Conventions

You awake one day in a new and strange place. You have no memories of how you got there, indeed, few or no memories. All you can tell is that you are having certain experiences: experiences that do not yet make much sense to you. You feel certain instinctive urges. You are confused and scared and you do not understand what is happening.

You are an infant.

We have all gone through something like this in the earliest stages of our lives. Our life begins as a submersion into the new and unfamiliar. What we are forced to do, to make sense of all of our experiences, is something like the following: from our experiences, we form perceptions; from these perceptions, we get impressions; and from these impressions, we form concepts. As an example, from a certain visual experience, I form an image (of a tree), the perception, from this image, I get certain impressions (it has a certain texture and color, it has various dimensions, some parts seem farther away than others, etc.). Finally, from all these impressions, I form the concept (of a tree standing before me).

Once we have concepts, what we need is some way to understand them, some way to handle them so as to get something from them. We thus embark on a project of organization and systematization. What we want, or what would be useful, is a way to categorize, generalize, simplify and synthesize these concepts into something meaningful.

This system is the seed of a language, which we use to handle the concepts we have obtained, and derive new concepts. Were we never introduced to formal, communicative languages (were we, for instance, stranded from birth on a desert island, what philosophers call a lifelong Crusoe), we might possibly develop a rudimentary language for our own use.

Shakespeare, through Juliet, famously said: “What’s in a name? that which we call a rose/ By any other name would smell as sweet”. We can take this to mean that a name is an arbitrary label we attach to a thing. The thing a name refers to is some set of particular set or class of objects, descriptions or designations: whatever its name, the thing remains constant, as the name itself is arbitrary and variable. In some sense, the name is almost abstract.

Every language has a system by which it attaches names to various concepts, and these names constitute the units of the language: the words. We call this system of naming the nominal convention of the language, which comprises a large part, if not most, of it. Let us call the entirety of all the concepts a person knows his concipulary, as contrasted with and similar to a vocabulary. A nominal convention is then a two-way mapping from the concipulary to a vocabulary: this concept of a tree corresponds to the word “tree”, and vice versa. However, not every concept might be mapped to a name, some concepts might be mapped to more than one name, and some names to more than one concept. It is then clear that for a nominal convention to be fully grasped, the user must have a certain concipulary as well, from which and to which to map. This instantiation of the requisite concipulary is one of the important aspects of learning a language, and learning in general.

Using a nominal convention has a great number of advantages and uses. It permits us to identify something and refer back to it: we label it, and can then keep it for later consideration. The name itself can serve as a mnemonic, as with onomatopoetic names. Names also allows us easily to refer to and handle various concepts, consider how they relate to each other, construct new ideas, and, importantly, open up the possibility of communication.

The rest of a language can be loosely broken up into two more components: a syntactical convention and an expressive/social convention. The syntactical convention delineates how the concepts are to be arranged into propositions, including grammar and punctuation. It describes how the various names and designations can be put together into a complete idea. These units of ideas, composed of concepts, are sentences or propositions. The syntactical convention is then the system by which words are formed into sentences, or by which concepts are formed into propositions. Note that a given nominal convention can have multiple syntactic conventions: for example, we can consider standard English and Yoda English (“John is a boy” vs. “A boy John is”).

Lastly, the expressive/social convention encompasses the various ways of expressing or conveying the ideas. This also covers the various social norms expected in communication. For instance, this would include spelling and pronunciation, tone and accent, but also social aspects, such as context-dependent vocabularies, honorifics, euphemisms, etc. Punctuation falls somewhere between this convention and the syntactical convention. Thus, this convention is an important component of language, as it constitutes the system whereby it is applied: it establishes the language as a communal effort. In addition, it establishes the language in the context of a culture, in which language has always played an important role.

Note that the various conventions have different roles in conveying meaning. The nominal convention specifies the referents in the communication, that is, what is being talked about. The syntactical convention specifies the content of the communication, that is, what relation or information is being conveyed. The expressive/social convention specifies how it is being communicated: what the context is and the means by which communication is taking place.

II. Communication

Communication is one of the main purposes of a language: language is the means by which the concepts in one head can appear in another head, so to speak. We will call the “speaking” party the sender and the “listening” party the recipient. There are essentially three components in communication: composition, transmission, and interpretation. Composition involves the translation of concepts by the sender into a message via a language. Transmission involves moving the message from the sender to the recipient. Interpretation involves the recipient translating the message back into concepts using a language. Note that the interpretation we speak of is only first level; it is only the translation of words into concepts and sentences into propositions and ideas, not the full extraction of implications. (For instance, “John washed his hands before lunch” could be interpreted to mean that John takes care to have sterile hands before having a meal, or it could be interpreted to mean that John had dirty hands sometime before lunch, or that John wanted to get something incriminating off his hands before a social occasion, etc. All the interpretation stage pertains to is the literal, first-level meaning of the sentence.) All of these steps are necessary and are found in any type of communication we recognize.

It is thus key that both communicators have a mutual agreement on the nominal, syntactical and expressive/social conventions being used. If the nominal convention is not fixed, the recipient may get a message in characters she recognizes, with a syntax that she can comprehend, but full of words that do not mean anything to her. If the syntactical convention is not fixed, the recipient may get a message with words she understands, in characters she recognizes, but in a structure she can’t understand or is ambiguous (“Tom saw Joe” means something different than “Joe saw Tom”). If the expressive/social convention is not fixed, the recipient may get a message in a way she can’t understand (mispronounced, written in strange characters, using obscure jargon, etc.).

The conventions being unspecified leads, then, mainly to difficulty for the recipient. If, however, the recipient is versatile enough, she may be able to infer some of the conventions (a new word, an unusual sentence structure, a new euphemism or honorific, etc.), but some background knowledge is essential, and a thorough knowledge of all the conventions makes communication much easier. We may here differentiate between coreferential communication and areferential communication. The former is distinguished from the latter by the use of a common reference or experience. For instance, in coreferential communication, the sender can point out something with just a demonstrative (“that thing in the sky” “that sound”), while in areferential communication, whatever is specified must be done so using only language (“the bright, yellow orb in the sky” “the harsh, ringing clang”). One of the goals of language is to move away from coreferential communication toward areferential communication. We will return to this later.

III. Errors and Obstacles to Communication

In attempts to communicate and achieve mutual understanding, many problems can arise, both in language and in communication. In a previous section, we divided communication into three stages: composition, transmission, and interpretation. Errors can occur at every stage, but there are even problems with the idea of communication.

The trouble with communication is that before communication can even take place, there has to be some base understanding of what will be communicated and how. For instance, before I tell someone something in English, there must be an implicit understanding that the communication will be done using spoken English (I can’t walk up to a Russian and begin speaking English, assuming that he will understand). Before I use a word, I have to assume that the word will be understood, and if the recipient does not, she may be able to infer the meaning from context, or else I will have to provide an explanation. This requisite preceding step we will call priming and poses some serious difficulties for communication, as it seems impossible to prime the conversation without communication. Since priming precedes communication, it seems priming must precede itself. As we will discuss later, priming serves as one of the main difficulties of communication. For now, we will assume priming has taken place.

An error in composition is one of the more trivial ones. It amounts to an incompetence on the part of the sender, in which the sender has either a malformed thought or has mistranslated that thought into language. A similar error can occur in interpretation due to incompetent translation of the message into concepts. This error is easily dealt with through diligence and care. We will thus henceforth assume both communicators are both competent once primed.

An error in transmission is simply where the message cannot get from sender to recipient either wholly or in part. It amounts to nothing more than an engineering problem: setting up an adequate system for relaying the message, intact, from sender to recipient. An error in this case would be a downed telegraph line, wax in one’s ear, a crummy postal service: something at least in principle correctable. There may be other, complex cases, but these are mere physical limitations. The problem is either physically irresolvable, which would preclude communication altogether, or it is resolvable, in which case a suitable apparatus would be sufficient.

Lastly, if priming is successful, the only thing that can impede interpretation is competence, and thus we can move straight into priming. Clearly, a proper priming will prevent any misunderstandings in the various conventions. But how can we prime at all? It seems that to prime, we need to communicate something, and to communicate anything, we first need to prime. So communication seems impossible, unless there is some sort of priming that comes standard, and allows further communication to be built up from it. As we shall see, there is indeed such a priming.

IV. Priming

As noted above, priming is the required step preceding any communication of establishing the various conventions that will be used. In effect, priming is the declaration of the language in which the rest of communication will be conducted. But how can we do this at all? I can’t say to someone who doesn’t understand English “this conversation will be conducted in English” as the recipient won’t even understand this establishing message: in order to understand the message, the recipient must already know that we are communicating in English, and, moreover, have an adequate knowledge of the English language.

So how can we prime at all? It seems that to prime, we need to communicate something, and to communicate anything, we first need to prime. Communication from the ground up thus seems impossible, unless there is some sort of priming that comes “built-in”, and allows further communication to be built up from it. As it happens, there is such a priming: it is the sort of fundamental understanding we would have of certain sorts of communication, which we shall call the natural priming. For instance, when someone points to something (or holds up an object, or draws a picture of something) and articulates a sound, we take that to mean that they are referring to the thing by the name given by the sound. Or, if the sender points to a certain sort of action, or imitates it, we can understand the name to refer to the action. The nominal convention is of gestures (gesturing in such and such a way means this), and seems innate: children and even animals seem to have such an understanding. The syntactical convention is merely in connecting the uttered noise or attached characters to the gesture.

The basis of natural priming is repetition and induction. Given that the sender does something repeatedly, the recipient infers that such an action means a certain thing. This is a sort of pavlovian response: the sender repeatedly says “tree” while pointing at a tree, and the recipient learns that, when the sender says “tree”, he is still referring to a physical tree, though he does not point to it. In essence, this is the installation of a certain nominal convention, whereby a meaning (the natural understanding of what is meant by pointing at a tree) is attached to a word (the association of the word/sound “tree” with the pointing to a tree).

Thus, clearly, this sort of communication is wholly coreferential: there must be some shared experience for establishing this sort of understanding. Moreover, this sort of priming is mostly if not only of the nominal convention, and is restricted by particular instances. For example, I can point to an object and say “tree”, but the recipient cannot understand that I mean “this, and everything like it in certain respects, is called ‘tree’,” but only “this particular thing is called ‘tree’.” We can perhaps ameliorate this by giving several different instances and using the same word to describe them all, thereby setting the stage for a generalization, but we cannot directly impart this generalization from examples. The generalization must be made by the recipient. Once a nominal convention basis has been established, the other conventions can be established pretty easily, either by imitation or instruction.

We can also establish a primitive basis, that is, a basis of concepts and their nominal designations derived from experience alone that themselves do not depend on other concepts. From this primitive basis we can construct other concepts and attach other names to them. This is the way language and its understanding develop, and it will feature importantly in our discussion of definitions.

Thus, as far as priming goes, if we have a reliable way to engage in coreferential communication, using nothing but natural priming, we can then build up communication to be quite widely useful. The goal is to be able to communicate areferentially, such that the sender and receiver can communicate with language alone. However, as we will see, even coreferential communication faces some problems.

V. Problems with Coreferential Communication

It seems one would think that this coreferential communication based solely on natural priming would be a sound foundation to begin establishing communication. We convey to the recipient the name of certain objects or actions by gestures, imitation or depiction, and thence to generalizations. This can be used to generate the primitive basis, from which higher levels of complexity and abstraction can be developed, leading to all concepts and all language. Everything seems to fall into place.

However, let us take a specific example: the sender points to a tree, says “tree” and the recipient from this learns that the thing pointed to is called a tree. But all this means is that the concept formed of the thing experienced has the same name according to both, but how can we be sure that both experiences are the same and that both concepts are the same? It certainly seem intuitive to think that they would be, and, lacking any reason to think otherwise, it seems reasonable to think they are, but how can we be sure? This is the classic problem of qualia, the individual instances of subjective experience. Two men can both agree that a thing is green (they both agree to call it that), but how can the one man be sure that the other man’s green is identical to his own? Perhaps what is red through the eyes of one is green through the eyes of the other, though the two both agree to give it the same name. Though it may in theory be possible to transmit or check qualia directly (some sort of device to transmit qualia as-is from one mind to another), it seems that it is a far cry in practice. This difficulty, which rarely causes problems, is thus correctable in theory, though intractable in practice.

But there is also trouble with the fact that each individual must (either by necessity or practicality), or at least often does, make her own generalizations or categorizations. A generalization or categorization is the procedure of taking a group of things all called by the same name and extracting common features so that new things can be included. This entails equating the categorization to a list of necessary and sufficient criteria, whether vague or specific. This set of criteria is used as a definition of what it means to be in that group, and will be very important in our later discussion of definitions.

However, often there is no specific set of criteria that is transmitted or even one that has widespread acceptance. This can allow even an individual’s understanding of a category to be vague: based on unknown or uncertain criteria. This often makes it difficult to discern the boundaries of a category (the necessary or sufficient conditions to qualify for inclusion). An example of this is the famous sorites problem of determining what a heap is. Namely, it asks what is the number of grains in the smallest heap. Other cases are the height of the tallest “short” man, or the hairiness of the hairiest “bald” man. This vagueness also leads to many disputes over border cases, in which the vagueness of common understanding allows for multiple conceptualizations that cover the same typical cases, but differ with respect to bordering cases. This shows why priming is so important, and also why it can be so difficult.

Some coreferential communication can be quite reliable in that we can pretty accurately check some concepts. It seems mathematical or geometric concepts can be generally well checked based on the exactness of what they correspond to and the precision that must hold for the many theorems that result from their application. In particular, numbers seem to be perfectly communicable (or at least nearly so), and thus anything that can be quantified can be perfectly well communicated as well. This plays into the importance of quantification in specificity, exactitude and understanding. But nevertheless there are, it seems quite many qualitative properties that cannot be exactly communicated.

VI. Definitions

As we have been discussing, prior to any communication, a stage of priming must take place, wherein the stage is set for communication. This involves specifying the various conventions (nominal, syntactical, expressive/social) such that the message sent by the sender can be properly understood by the recipient. A very important part of priming is the introduction of definitions, that is, explanations in other terms of what a word or term means, what it will be used to mean, or how it will be used. We will call the term being defined the definiendum, and the collection of other terms used to define it the definiens. The terms used in the definiens we will call subterms (note how even now I am priming the discussion by introducing definitions!). There are various sorts of definitions that we will discuss, but to begin, we will introduce a basic theory of definitions.

For a definitions to be useful, all the subterms in the definiens must be understood: if we know the subterms, we can know the new term, given the definition. Thus, for the definition to be useful, its subterms cannot include the definiendum (as then we would need to know the term before we could understand the term, which is obviously not possible). Sometimes even subterms will need to be defined, and so on. But clearly this cannot go on forever: there must be some point of understanding at which we can stop. If not, then either this presents a vicious infinite regress, wherein to understand a term we first need to understand another term, and so on ad infintum, or becomes circular, wherein we need to understand A to understand B, B to understand C, and C to understand A.

Thus, there must be terms that do not require other terms to be understood. These are called primitives or primitive notions, and serve as the axioms of the definition system (examples typically given are color, being, number, and duration). Typically, however, it is unnecessary to reduce all definitions to the level of primitives, and so we merely must reduce the definitions to the level of acceptability: the point at which the meaning of the terms is mutually accepted as understood. It is possible that two conversers could hold some mutual level of acceptability and then a third comes along with a lower level and definitions would need to be added to get everyone on the same page (this often happens when a new person is introduced into a technical field). In whole, our terminology forms, as it were, a pedigree, with each of the various terms tracing its lineage back to some combination of primitives.

Let us now distinguish between various sorts of definitions. The classic distinction is between intensive and extensive definitions. An intensive definition gives necessary and sufficient conditions: what, categorically, would include something under the term. For instance, we can define a “square” to be “a closed planar figure of four congruent straight line segments and all angles equal”. In contrast, rather than provide the conditions for inclusion and exclusion, an extensive definition gives an exhaustive list of all the things in the specified category. For example, we can define a continent to be any one of the landmasses Africa, Asia, Europe, Antarctica, Australia, North America, or South America. Ideally, an intensive definition is always preferred, as it allows us to understand what it means to be in a category, and is in no way contingent upon extant examples. A third kind of definition is called an ostensive definition which is formed through coreferential communication. Essentially, it is what we have already described: we gesture toward an object and say a name and thereby define that name by that object (or name that object by that word).

We may also distinguish between definitions with different functions. The first is stipulative definition, in which a term is applied to a novel concept as a way to refer to it. This is typical in such fields as mathematics and philosophy (“stipulative definition”, for instance, is here given a stipulative definition). Another sort is a descriptive definition, which attempts to clarify or specify a common term that does not have a robust definition. This sort of definition can be disproven by showing that there is something the definition includes or excludes that the common notion does not. A precising definition is something of an intersection of the two, in which we take a subset of what is commonly considered part of the meaning of a term and specify that we will use the term to mean that subset. For instance, though “ball” can often be used to describe many sorts of round objects (e.g. footballs), we can say that, in our discussion, we will only use them to refer to the spherical variety.

The general form of a definition is a genus with differentia, that is, the class to which it belongs and what distinguishes it within that class. For example, we earlier defined a square as “a closed planar figure of four congruent straight line segments and all angles equal”. In this case, the genus is closed planar figures, and the differentia are that it has four congruent straight sides. We can also define terms as the intersection of genera. For instance, we can define squares as the intersection of regular polygons and quadrilaterals, or as the intersection of rectangles and rhombi, etc. (this is still basically using the genus/differentia approach, as the differens in this case would be the inclusion in the other genus) In this way, we can keep generalizing, using classes that take more and more elements, but we will somewhere need to stop, and this is arguably the point at which basic concepts come in (such as “entity” or “class”, etc.).

In forming definitions, there are several precepts that serve as generally if not universally helpful. As we have discussed, whenever possible, it is best to define terms using simpler or more well-understood subterms. If the subterms are as complex or obscure (or more complex or obscure) than the definiendum, then the definition is clearly counterproductive, and doing so is said to be obscurum per obscuris ([clarifying] the obscure through the obscure). The definition should not be circular, either individually, or globally. The definition should be intensive wherever possible (that is, it should delineate the essence of the definiendum). The definition should be adequate, in the sense of including all the things to which the term applies, and exclude all the things to which the term does not apply. It is typically preferable, where possible, not merely to recast a definition in terms of a synonym or the negation of an antonym.

As communication and discussion hinges on mutual understanding, defining terms is always necessary (unless the terms fall into the “acceptable” category already discussed). If mutual understanding is the goal, definitions are indispensable. Thus, in discussion, we should always be able and willing to define our terms down to the acceptable level if needed, and we should preemptively define new, vague, or unusual terms, and communicate how they will be used. A failure to do so is a failure to prime the communication, and leaves communication gravely at risk.

VII. Problems with Poorly Primed Communication

As we have seen, priming is an essential prerequisite for successful communication. Priming mainly involves, in practice, introducing the requisite definitions (mainly stipulative, descriptive and precising) so that both communicators are on the same page. But too often it happens that priming does not take place, or it is inadequate (or even that it is impossible or impracticable), and this can lead to serious misunderstanding and miscommunication.

As we have mentioned, an important part of priming is the fixing of the nominal convention that will be used, which is mainly done through definitions. Also as we have mentioned, the nominal convention is the two-way mapping between the concipulary (set of concepts) and the vocabulary (set of names). A definition is merely the introduction or specification of the correspondence of one concept to one name and vice versa, that is, a word (this may be a nonstandard usage, in that “word” typically refers to the name with potentially multiple meanings, but I will use it to mean a name-meaning pair. Thus “can” meaning a “cylindrical metal container” is a different word than “can” meaning “be able to”).

Let us examine the case of one word used in communication: the sender has a concept, attaches to it a name, and sends this (via some medium) to the recipient, who takes the name and finds what concept it maps to in her vocabulary. All we will assume is that the two communicators are not using the same nominal convention. Let us call the concept of the sender’s word the intended concept. We can distinguish five cases, not all of which are mutually exclusive: (1) recipient connects the name to the intended concept (2) the recipient connects the name to a different concept (3) the recipient connects the name to no concept (4) the recipient attaches a different name to the intended concept (5) the recipient does not have the intended concept in her concipulary.

Case (1) is clearly a case of successful communication, priming notwithstanding.

Case (2) is a conceptual confusion, in which a name is mistaken to mean something other than the intended concept. This is a language glut: the same name is doing too much. We will call such a case a homonymopathy (a mistaken use of the same name to mean something unintended: the name is a pathological homonym).

Case (3) is a nominal failure, in which the recipient receives what is to her nonsense. Perhaps she then tries to assign it a concept by guessing, in which case it would become (1) or (2) if she guessed rightly or wrongly respectively.

Case (4) is a nominal confusion, in which, had the user used a different name, communication could have been successful. This may easily be combined with the preceding three cases, but the point is that communication was unsuccessful because of a poor choice of words. This is a language gap: the same concept can be reached, but the two users are separated by different names for it (both an Englishman and a Frenchman know what water is, but “water” means nothing to one and “eau” means nothing to the other). We will call such a case a synonymopathy (a mistaken attempt to communicate the a concept by an unknown name: the two names are pathologically synonymous).

Case (5) is merely a case of conceptual failure: priming and education must precede in order to get the recipient to apprehend the intended concept, much less recognize it by name.

We thus see that the five cases amount to one success, two failures and two confusions. The success obviously needs no mending, and the failures are ameliorable, in that they serve to halt communication, not lead it astray. However, even a synonymopathy can halt communication, so the major troublemaker is a homonymopathy. As I will later argue, I think homonymopathy is one of the greatest impediments to successful communication.

As discussed, nominal and conceptual failures can be remedied by definitions and explanations respectively. But what about nominal and conceptual confusions? The mythical case of the tower of Babel in the Old Testament is a clear case of nominal confusion: in order to halt the progress of the construction of a tower, god says “let us go down, and there confound their language, that they may not understand one another's speech.” That is, formerly, they had well-synchronized nominal conventions, but god perturbs these conventions, so that, though they are referring to the same thing (the building of the tower) they cannot understand one another. Thus, we may say that a name for such a case of nominal confusion is a Babel, or a Babel phenomenon. The case of conceptual confusion is somewhat similar, and so we will give it the inventive name of inverse Babel, or an inverse Babel phenomenon. Again, an inverse Babel is a case in which the users use names that one another recognizes, but the meaning to one is different than the meaning to another. For example, both an Englishman and Frenchman recognize the word “pain”, but the meaning to one is not the meaning to another: if one said “pain” to another, the second might give the first a baguette when he wanted an aspirin.

Before we address solutions to each, it is important to identify how we can recognize each sort of confusion. A Babel can be identified if two people point to the same things and say something different, or if the explanation for the term in other language is the same “here, we call that…”. Another, simpler way is if we have some background knowledge: if you are an architect and you go to a painting conference in a foreign country, you will expect the people there to be discussing architecture, even if you don’t know what their language means when you encounter them. Much more simply, if you believe the sender is competent, but can’t understand what he is sending you, there is probably a language gap (though there may also be a conceptual failure). The ways to correct for this are well-studied and well-documented (it basically involves developing a method of translation, or choosing a standard nominal convention to use). It is possible that the nominal conventions are so dissimilar that coreferential priming is required.

An inverse Babel is much trickier to discern. Typically the sender will send you something but the way he uses a term will be jarring or strange. This can be verified by asking “what do you mean?” and seeing if his reply matches your understanding. The lower down in the conceptual hierarchy the homonymopathy lies, the more difficult it is to remedy: this was raised in the section on problems with priming and coreferential priming. It may be the case, as with a vague or contested concept, that his concept and yours will be similar, but nevertheless, you will be using the same word to mean different things. Sometimes a loose, imperfect definition will do, as we have discussed, but one should not expect these vague conceptions to overlap everywhere. In general, once a homonymopathy has been identified, all that is needed is to assign each concept a different, mutually agreed-upon name. If your conception of a “pile” is different than mine, we need not call them by the same name.

We can here identify a sort of fallacy which I will call the nominal fallacy (note that there is already a fallacy of this name but of different meaning). One commits a nominal fallacy when one thinks that two words must mean different things because they have different names (synonymopathy) or two words must mean the same thing because they have the same name (homonymopathy). The fallacy can be easily demonstrated by showing that “Little” means the same thing as “small” and “bow (archery)” has the same name as “bow (knot)”. Though this fallacy can be easily addressed, it can also be quite tenacious, especially when the difference in meaning is slight or when personal stock is put in the names.

Priming is a vital and important step which cannot be foregone. Failing to prime adequately leads to such problems as synonymopathies and homonymopathies, in which the communicators fail to use language the other understands, or worse, uses language the other thinks he understands. Avoiding these takes care but whatever the cost in preventing them greatly outweighs the hazards of ignoring them. In the final section, we will look at the relevance and applications of the preceding discussion.

VIII. Relevance and Applications to Discussion

Discussion and dialogue depend upon communication. Communication, as we have been discussing, depends crucially on having both communicators first agreed on the conventions that will be used, particularly the nominal convention. If communicators fail to prime the conversation by establishing the requisite conventions, the discussion can devolve into synonymopathies and homonymopathies, such that the communicators are merely speaking past one another. In many areas, these can manifest as major controversies.

Armed with our previous discussion, we can distinguish two different sorts of disputes that arise: terminological disputes: disputes over the meaning and use of terminology; and factual disputes: disputes over what is the case, as described by the agreed-upon terminology (some may add to this other disputes, such as normative disputes). Clearly any (or at least most) terminological dispute can be dealt with through proper priming and clarification, or at least any dispute arising.

Philosophy is the study of ideas generally, though even this definition is not uncontroversial. It is known more for providing questions and perspectives than hard answers, for broadening perspectives rather than narrowing them. Indeed, there is just about no area of philosophy that is not at least somewhat controversial. Philosophy is also known as being home to many disputes which have persisted since antiquity. I hypothesize, however, that many of these disputes are terminological, perhaps even a majority, or at least arise therefrom.

Vagueness is a common source of terminological dispute. Two people have two different criteria for what qualifies for inclusion in a certain category, with the vast majority of cases mutually agreed upon. However, certain marginal cases are included in one but excluded in another. Each thinks he is correct, and in a sense both are: they are just speaking about different things—things that are very similar, and yet subtly different. There is no shortage of vague or essentially contested concepts in philosophy: knowledge, self, consciousness, explanation, cause, abstract, existence, good, obligation, nature, reason, beauty, purpose, essence, time, meaning, experience, person, to name a few. Vagueness also breeds uncertainty, which can be loosely characterized as a dispute with oneself: part of you thinks it should be interpreted one way, another part thinks it should be another way. If we are to resolve both disputes and uncertainty, we must find a way to eliminate or at least reduce vagueness.

We can thus suggest ways in which to present and defend an argument. Before anything, the first step must be to define the relevant terms initially, or else be careful to define them as they come up. Definitions cannot in practice be perfect, but one must endeavor to make them as clear and complete as necessary and possible. If one is presenting an argument, one define to acceptability all important terms, making sure to limit the senses of vague terms especially. If one is rebutting an argument, one must examine the possible senses in which the terms of one’s opponent’s argument can be taken. If there are multiple, interacting terms, for charity, one must examine the various combinations of meanings and address each. This is especially important in order to avoid and demonstrate equivocation: one must be sure that the terms are being used in the same way throughout the argument, or else that the various meanings all conspire to reach the same conclusion. If one is merely engaged in a less argumentative discussion, one must be sure to consider, evaluate and decide on the terms before the discussion has proceeded too far. It is no use discussing something for hours only to find out that the entire point could have been avoided had you recognized an inverse Babel early on.

This is perhaps the most important point: fixing definitions is an indispensable part of any discussion. Sometimes, an entire discussion can be wholly or mostly forgone if the definitions are made clear at the outset. I am convinced that a significant part of current philosophical discussion results from little more than vagueness and homonymopathy. It may be just as difficult to get another to understand what you mean as it is to get another to agree with you. What is most needed is clarity and mutual understanding: conversation must always precede discussion.

IX. Glossary

Lifelong Crusoe
    A person who, for his whole life, has never come into contact with others or society.
    The set of names in a certain language an individual can recognize as corresponding to certain concepts or which an individual can use to convey certain concepts.
    The set of concepts an individual understands, can recognize as corresponding to certain names in a language, or can convey using names in a certain language.
Nominal Convention
    An agreed upon system of attaching names to various concepts (and concepts to certain names) which comprises one of the main components of a language.
Syntactical Convention
    An agreed upon system of forming sentences from words, that is, forming propositions or ideas from concepts. This typically involves a grammar, punctuation, syntax and various constructions.
Expressive/Social Convention
    An agreed upon system of expressing ideas, typically including situation- dependent vocabulary (e.g. honorifics, jargon, complexity level) methods of communication (e.g. spelling, script, diction, enunciation, accent, intonation).
Steps of Communication:
    The step of communication that involves the sender translating the idea to be communicated into a message via a language.
    The step of communication that involves moving the message from sender to recipient.
    The step of communication that involves the recipient translating the received message into ideas via a language.
Coreferential Communication
    Communication in which there is a common reference or experience that both communicators can directly experience.
Areferential Communication
    Communication in which there is no common reference or experience that both communicators can directly experience.
    The necessary prerequisite for successful communication that involves establishing the various conventions that will be used.
Natural Priming
    Priming one has naturally that allows one to infer meaning by gestures, repetition or imitation.
Primitive Basis
    The fundamental concepts which themselves cannot be formed or described using more basic concepts, generally derived from raw experiences or even a priori.
    A definition that specifies the necessary or sufficient conditions for qualifying for inclusion.
    A definition that gives an exhaustive listing of all things that qualify for inclusion.
    A definition in which, by pointing to a certain thing or set of things (either figuratively or literally), one either gives instances which can be generalized or determines what qualifies for inclusion.
    A definition in which one specifies by assertion what a term means or will be used to mean.
    A definition in which one attempts to give a meaning to a term that reflects how it is commonly understood.
    A definition in which one narrows the scope of a term from how it is commonly understood to reflect how it will be used.
    The word or term to be defined.
    Whatever is used to define a word or a term, typically composed of a set of composed subterms.
Primitive (notion)
    Terms that do not require other terms to be understood that function as axioms of conceptual understanding.
    The super-category into which the thing being defined falls.
    The property that sets apart the thing being defined from other things in its genus.
    One of the terms in the definition of a term.
Obscurum per Obscuris
    Defining a term using more difficult to understand subterms.
    A failure of communication in which a term with an intended meaning is mistaken to mean something else instead.
    A failure of communication in which a term is not understood to have the intended meaning, but a different term could have been used to achieve an understanding of the intended meaning.
Babel (phenomenon)
    A case in which communicators are trying to communicate the same meaning using different terms or languages.
Inverse Babel (phenomenon)
    A case in which the same term is interpreted to mean different things by different communicators, or in which the same term is being used in different ways.
Nominal Fallacy
    Any of the following fallacious lines of reasoning (X and Y being words or terms):
    1. X and Y have different names.
    2. Therefore X and Y refer to different things.
    1. X and Y refer to different things.
    2. Therefore X and Y have different names.
    1. X and Y have the same name.
    2. Therefore X and Y refer to the same thing.
    1. X and Y refer to the same thing.
    2. Therefore X and Y have the same name.
    (Not to be confused with a different fallacy of the same name).