Associative Nets

I am most grateful to Brian Kennedy for his detailed and insightful reply to my earlier blog Are there infallible facts. I’ve already followed up with the response Empirical vs. Ex Cathedra Solutions. Here is a second follow-up, selecting other points made by Brian, which certainly have taken the discussion beyond the limits of my earlier piece.

Brian pointed out that I had “exposed… that the noun *claims* is (often) subjected to more weight than it can bear”, but he also rightly pointed out that this is quite common in everyday language. He also makes the valuable point that ordinarily words reach over to connected with others, as when the term *claim* is associated with claimants, liabilities, and assets (especially in a legal context), although it is quite acceptable and common to use each of these terms without making explicit reference to any of these. Elsewhere — and more recently — I referred to this as cases where words are part of an associative-net.

brain-netWe can, for example, ask others to “free associate” by giving a starter-word only and asking someone to come up with as many words as possible for the next few minutes. In many cases the string of words each person presents during such serial association have significant overlaps. Wittgenstein talked about this phenomenon as terms having a “family resemblance”. It underscores that words or phrases should NOT be viewed in the manner suggested earlier, as independent items — a view proposed, for example, by an earlier Wittgenstein (c.1921, the author of the Tractatus) and also by members of the Vienna Circle (later knows as Logical Positivists), but that words reaches out to others, as it were (see Wittgenstein post-1940).

For example, we understand that *claim* as used by lawyers and accountants is viewed as something owed (liability) or owned (asset). Furthermore that where there is a claim there is also a claimant, a party making such a claim, whether as an individual or a group of people. Thus there have been two greatly opposed positions. One states that each word should be viewed as a separate, independent event, an item which is clearly demarcated from others. The second position argues that words, in general, are members of families, have resemblances and acted on behalf of other members of their family. A useful analogy to those of us who have families!

I generally take the position that in ordinary language we are best advised to assume the “family resemblance” stance. Most referential words and phrases follow the prescription that there is more than one meaning to a words and/or phrase, as demonstrates most clearly in any etymological dictionary, or in Roget’s Thesaurus. In current-day formal sciences, however, the rule is that each word or phrase should have a rigid definition (although this rule is often broken within a relatively short time). These two rules clearly conflict. But it is like a tennis player on a badminton court: he either has already learned separate sets of rules and uses these appropriately, as the occasion demands or, if he fails to do so, he will surely be asked to leave the court by some imperious judge!

Let us briefly look at a family of adverb-adjectives which are often reified and which thereby get transformed during this process. Thus *truly*, *true* and *truth* (see example 1 below) represents such a family: these items have family resemblance, i.e., they are related in meaning, but have different grammatical status. However *truth* — the third of these — overarches, literarily speaking: it is an abstract entity relative to the other two terms, and “sits above” them, as it were.

Example 1
Item: truly, true, truth
Over-arching: truth

Example 2
Item: factually, factual, fact
Over-arching: factuality

The further examples that follow are similar but include terms often used in philosophical discussions and are “noun-words”, which appear as if these are names of objects — which they are not.

  • really, real, reality
  • necessarily, necessary, necessity
  • logically, logical, logic
  • infallibly, infallible, infallibility

Why say, “facts [as items] are claims to truth [in the over-arching sense]”? Why not say instead, “truths [items] are claims to factuality [over-arching]”?

How is the *facts/factuality* cluster to be distinguished from the *truths/truthfulness* cluster? Which of these two clusters (if either) is logically/definitionally dependent on the other? In what do these dependencies consist? Do these two clusters constitute a distinction without a difference?

Furthermore, whatever a decision is reached in these matters, one should remember that these have a limited time during which they operate reliably. Our efforts to catch the world as “experienced”, on the fly, is a game with rules that we now know changes as we play.*

*Imagine playing a set of tennis during which the rules change! This has mattered little in the past, when games were long and players generally speaking did not outlast the set, as they do now. And don’t let’s even get started in games which we believe will continue “for all time.”

Facts and the Web of Conceit

Part 1 — Corroboration of Beliefs

When people talk about themselves — about what they saw, how they felt about everyday things, even how they discuss unusual happenings — their utterances could often have been expressed more succinctly, more accurately, with more of a twinkle in their eyes or fervour in the voice. I myself invariably look for “better” ways of describing things, especially personal impressions and experiences. I am acutely aware of the difference between how I think about past experiences and how I report these to others. There is a conflict between the best words which give the most accurate description and to tailor the suit to the wearer.

Sometimes a matter of making things interesting rather than boring. One may take a measure of the other person, of their sensibilities and their sophistication. But one also needs to draw a line,to reach conclusions readily, correctly, avoid dilly-dallying, tell it as it comes even when it is not how things originally appeared, by a good preacher, not a yawning lecturer. Such doubts mostly concern personal experiences and do not apply to reporting “objective” matters.

Thus when I tell you that I flew from Toronto to London on the 8th September, 2011 this report that can be independently supported, that is, corroborated by independent “others”. Corroboration by others supports any beliefs I express on my own behalf.*

Belief, we can conclude, is either something we confirm to be “true” because we trust our personal experience (even when we shouldn’t!) or because we trust others regardless of whether we trust ourselves in a matter. (Was I awake or dozing when I thought I saw a mouse climb up a clock?)

Belief is therefore not a property or quality of an event, but refers to a personal, subjective state to which we have assigned transcendental qualities, namely a quality which cannot be independently verified. We do so on the assumption that it is an act of affirmation which others cannot deny us — even dare not deny us!

When used in the plural, e.g. (a) my beliefs, suggesting that I have several beliefs which together form an entity, often referred as a set of beliefs or that (b) our beliefs, that is a set of individual beliefs each of which is also part of each person’s set of beliefs, there is a problem.

Take each in turn: (a) my beliefs suggests that I am the owner of several belief-items, that each item is, or may be independent of others and that these have no common origin. It is like having individual pearls in my jewellery box, not strung together into a necklace or fashioned into an ornament. Furthermore some pearls may have a different hue than others, or differ in size, features not mentioned or discussed by my statement.

828002.TIFBut beliefs, like pearls, differ in brightness, size, and other properties and some beliefs are more intensely believed than others. Also different beliefs held may be coherent, that is, constitute a web of beliefs and therefore have a demonstrable structure which relate them to each other or which like pearls lie separately yet waiting to be arranged into one or more pieces of jewellery.

It seems more often that the latter describes things best, even though we fantasize otherwise on the assumption that beliefs should form a “configuration”. We may occasionally be able to demonstrate that some belief-items cohere though the wish here may be father to the claim.

Consider the term (b) our beliefs (which I treat as one word referring to one item). This may refer to one of two notions: first, that each person has a set of (independent) foundational ideas, which is referred to as “our beliefs” since it refers to an individual person’s set of individual beliefs. Yet my set of beliefs and yours may differ, so that it it is appropriate to refer to these in the plural form, as our beliefs, your set and mine.

Note, that belief-items may be shared, but that each set is different — that individual belief-item is not necessarily shared (common) by all members of the group of people. It may result in the group being separated into sub-groups, as are those members of a church who hold that only men should be priests, or who hold that only unmarried men should be admitted to priesthood — currently a debating point in several religious communities.

Most often we agree to sort beliefs into those we hold in common with others to form a belief-cluster (or dogma?) which unites us into a group and which therefore leave beliefs which lies outside this shared area as “idiosyncrasies”. It often represents a first step towards open-mindedness. United we stand, cry some! Let anthropologists and sociologists characterize us as best they can.

If we ask, “Can a list of ideas everyone shares with others be given a basket-label?” we come up with two answers, not one! There often is a class of foundational ideas but there is often also a basket of transcendental ideas. Foundational ideas are those which refer to experiences which cannot be destroyed by doubt, ideas which are skeptic-proof.

Descartes (c. 1630) comes to mind as the best known advocate of this position, but also Locke (c. 1690) who claimed that all ideas were experience-based and mediated initially through sense-perception: the slate for Locke was clean but also ready to receive messages, unmistakeable imprints.†

Accordingly, foundational ideas refer to those first impressed (imprinted?) on us and which therefore cannot be eradicated or revised, but only be built upon. Some writers have argued that first-impressed does not mean that what subsequently follows is inevitable since there is no rule which states that such material has subsequently to be processed in a particular manner. Such a claim would be like claiming that given flour, water and a few basic ingredients one can only create one type of bread and no cakes. Tell that to a lover of apple strudel. One may say that the structure of the web is not determined by what the web is composed of — as any graphic artist knows. The relationships between materials used to make a painting is only one of many possible relations.

It is widely held that transcendental ideas are those we use to order our experiences. According to Kant (c. 1800) ordering experience requires concepts of “time” and “space” yet these are entities which are independent of the experience that is being ordered, as are the logical categories which are part of the ordering process. The existence of such ordering ideas have always been presumed as givens (Plato wrote extensively about these).

In principle it would be possible to give each experience its own unique name. Nobody knows how this could work. I see a sheep now and I could assign it a sheep/to-day’s date/present clock-time tag! It is a problem in coding and in handling the codes and book-keeping. The next sheep I encounter would carry the name sheep/current date/present clock-time tag. No person I know would be comfortable in such a world, but would invent short cuts — but machines could function quite well in such a world. In other words, it is possible to conceive and to build devices which operate according to such sets of rules.

We don’t know — before plans are actually laid out — how effective and functional such a device would be, or what it would have to look like. At first glance it may work quite well: known as a robot it would operate according to emerging principles of robotics. I don’t know whether robots have to “feel comfortable”, but I suspect not. Robots do not have beliefs in our earlier sense. Vive la difference. Without beliefs they may not go to war!

A robot then would operate according to emerging principles in robotics. Nevertheless robots could change their actions and adapt to their environment — including their self-made environments — but the operating principles need to be worked out in advance — by us, we who are their originators or creators.

*Footnote: I have written a separate blog on corroboration which will be published soon.

†Footnote 1: There is a modern school of thought which reversed this position and holds that all ideas are influenced by the existence of other ideas concurrently in circulation. Contextualism has its roots in Charles Pierce’s philosophy of pragmatism and its most notable advocate in mid-20th century was W. Quine (c. 1950-1990).
Footnote 2: Contextualism was inherently unfriendly to logical empiricism, as publicized earlier by A.J. Ayers (1936) in his widely influential book
Language, Truth and Logic, a book which echoed many of the positions first espoused by the early L. Wittgenstein and subsequently by R. Carnap and members of the Vienna Circle and after the Anschluss in 1938 by readers of the journal Erkentniss and its USA successors).