A teleological explanation is one that explains things and features by relating to their contribution to optimal situations, or to a normal mode of functioning, or to the attainment of goals by a whole or by a system to which the said things or features belong. It often involves the confusion or reversal of causes and effects and the existence of some "intelligence" at work (either self-aware or not).
Socrates tried to understand things in terms of what good they do or bring about. Yet, there are many cases when the contribution of a thing towards a desired result does not account for its occurrence. Snow does not fall IN ORDER to allow people to ski, for instance.
But it is different when we invoke an intelligent creator. It can be convincingly shown that intelligent creators (human beings, for instance) design and maintain the features of an object in order to allow it to achieve an aim. In such a case, the very occurrence, the very existence of the object is explained by grasping its contribution to the attainment of its function.
An intelligent agent (creator) need not necessarily be a single, sharply bounded, entity. A more fuzzy collective may qualify as long as its behaviour patterns are cohesive and identifiably goal oriented. Thus, teleological explanations could well be applied to organisms (collections of cells), communities, nations and other ensembles.
To justify a teleological explanation, one needs to analyze the function of the item to be thus explained, on the one hand and to provide an etiological account, on the other hand. The functional account must strive to elucidate what the item contributes to the main activity of the system, the object, or the organism, a part of which it constitutes, or to their proper functioning, well-being, preservation, propagation, integration (within larger systems), explanation, justification, or prediction.
The reverse should also be possible. Given information regarding the functioning, integration, etc. of the whole, the function of any element within it should be derivable from its contribution to the functioning whole. Though the practical ascription of goals (and functions) is problematic, it is, in principle, doable.
But it is not sufficient. That something is both functional and necessarily so does not yet explain HOW it happened to have so suitably and conveniently materialized. This is where the etiological account comes in. A good etiological account explains both the mechanisms through which the article (to be explained) has transpired and what aspects of the structure of the world it was able to take advantage of in its preservation, propagation, or functioning.
The most famous and obvious example is evolution. The etiological account of natural selection deals both with the mechanisms of genetic transfer and with the mechanisms of selection. The latter bestow upon the organism whose features we seek to explain a better chance at reproducing (a higher chance than the one possessed by specimen without the feature).
Hitherto, we have confined ourselves to items, parts, elements, and objects within a system. The system provides the context within which goals make sense and etiological accounts are possible. What happens when we try to apply the same teleological reasoning to the system as a whole, to the Universe itself? In the absence of a context, will such cerebrations not break down?
Theists will avoid this conundrum by positing God as the context in which the Universe operates. But this is unprecedented and logically weak: the designer-creator can hardly also serve as the context within which his creation operates. Creators create and designers design because they need to achieve something; because they miss something; and because they want something. Their creation is intended (its goal is) to satisfy said need and remedy said want. Yet, if one is one's own context, if one contains oneself, one surely cannot miss, need, or want anything whatsoever!
III. The Issue of Context
If the Universe does have an intelligent Creator-Designer, He must have used language to formulate His design. His language must have consisted of the Laws of Nature, the Initial State of the Universe, and its Constants. To have used language, the Creator-Designer must have been possessed of a mind. The combination of His mind and His language has served as the context within which He operated.
The debate between science and religion boils down to this question: Did the Laws of Nature (the language of God) precede Nature or were they created with it, in the Big Bang? In other words, did they provide Nature with the context in which it unfolded?
Some, like Max Tegmark, an MIT cosmologist, go as far as to say that mathematics is not merely the language which we use to describe the Universe - it is the Universe itself. The world is an amalgam of mathematical structures, according to him. The context is the meaning is the context ad infinitum.
By now, it is a trite observation that meaning is context-dependent and, therefore, not invariant or immutable. Contextualists in aesthetics study a work of art's historical and cultural background in order to appreciate it. Philosophers of science have convincingly demonstrated that theoretical constructs (such as the electron or dark matter) derive their meaning from their place in complex deductive systems of empirically-testable theorems. Ethicists repeat that values are rendered instrumental and moral problems solvable by their relationships with a-priori moral principles. In all these cases, context precedes meaning and gives interactive birth to it.
However, the reverse is also true: context emerges from meaning and is preceded by it. This is evident in a surprising array of fields: from language to social norms, from semiotics to computer programming, and from logic to animal behavior.
In 1700, the English empiricist philosopher, John Locke, was the first to describe how meaning is derived from context in a chapter titled "Of the Association of Ideas" in the second edition of his seminal "Essay Concerning Human Understanding". Almost a century later, the philosopher James Mill and his son, John Stuart Mill, came up with a calculus of contexts: mental elements that are habitually proximate, either spatially or temporally, become associated (contiguity law) as do ideas that co-occur frequently (frequency law), or that are similar (similarity law).
But the Mills failed to realize that their laws relied heavily on and derived from two organizing principles: time and space. These meta principles lend meaning to ideas by rendering their associations comprehensible. Thus, the contiguity and frequency laws leverage meaningful spatial and temporal relations to form the context within which ideas associate. Context-effects and Gestalt and other vision grouping laws, promulgated in the 20th century by the likes of Max Wertheimer, Irvin Rock, and Stephen Palmer, also rely on the pre-existence of space for their operation.
Contexts can have empirical or exegetic properties. In other words: they can act as webs or matrices and merely associate discrete elements; or they can provide an interpretation to these recurrent associations, they can render them meaningful. The principle of causation is an example of such interpretative faculties in action: A is invariably followed by B and a mechanism or process C can be demonstrated that links them both. Thereafter, it is safe to say that A causes B. Space-time provides the backdrop of meaning to the context (the recurrent association of A and B) which, in turn, gives rise to more meaning (causation).
But are space and time "real", objective entities - or are they instruments of the mind, mere conventions, tools it uses to order the world? Surely the latter. It is possible to construct theories to describe the world and yield falsifiable predictions without using space or time or by using counterintuitive and even "counterfactual' variants of space and time.
Another Scottish philosopher, Alexander Bains, observed, in the 19th century, that ideas form close associations also with behaviors and actions. This insight is at the basis for most modern learning and conditioning (behaviorist) theories and for connectionism (the design of neural networks where knowledge items are represented by patterns of activated ensembles of units).
Similarly, memory has been proven to be state-dependent: information learnt in specific mental, physical, or emotional states is most easily recalled in similar states. Conversely, in a process known as redintegration, mental and emotional states are completely invoked and restored when only a single element is encountered and experienced (a smell, a taste, a sight).
It seems that the occult organizing mega-principle is the mind (or "self"). Ideas, concepts, behaviors, actions, memories, and patterns presuppose the existence of minds that render them meaningful. Again, meaning (the mind or the self) breeds context, not the other way around. This does not negate the views expounded by externalist theories: that thoughts and utterances depend on factors external to the mind of the thinker or speaker (factors such as the way language is used by experts or by society). Even avowed externalists, such as Kripke, Burge, and Davidson admit that the perception of objects and events (by an observing mind) is a prerequisite for thinking about or discussing them. Again, the mind takes precedence.
But what is meaning and why is it thought to be determined by or dependent on context?
Many theories of meaning are contextualist and proffer rules that connect sentence type and context of use to referents of singular terms (such as egocentric particulars), truth-values of sentences and the force of utterances and other linguistic acts. Meaning, in other words, is regarded by most theorists as inextricably intertwined with language. Language is always context-determined: words depend on other words and on the world to which they refer and relate. Inevitably, meaning came to be described as context-dependent, too. The study of meaning was reduced to an exercise in semantics. Few noticed that the context in which words operate depends on the individual meanings of these words.
Gottlob Frege coined the term Bedeutung (reference) to describe the mapping of words, predicates, and sentences onto real-world objects, concepts (or functions, in the mathematical sense) and truth-values, respectively. The truthfulness or falsehood of a sentence are determined by the interactions and relationships between the references of the various components of the sentence. Meaning relies on the overall values of the references involved and on something that Frege called Sinn (sense): the way or "mode" an object or concept is referred to by an expression. The senses of the parts of the sentence combine to form the "thoughts" (senses of whole sentences).
Yet, this is an incomplete and mechanical picture that fails to capture the essence of human communication. It is meaning (the mind of the person composing the sentence) that breeds context and not the other way around. Even J. S. Mill postulated that a term's connotation (its meaning and attributes) determines its denotation (the objects or concepts it applies to, the term's universe of applicability).
As the Oxford Companion to Philosophy puts it (p. 411):
"A context of a form of words is intensional if its truth is dependent on the meaning, and not just the reference, of its component words, or on the meanings, and not just the truth-value, of any of its sub-clauses."
It is the thinker, or the speaker (the user of the expression) that does the referring, not the expression itself!
Moreover, as Kaplan and Kripke have noted, in many cases, Frege's contraption of "sense" is, well, senseless and utterly unnecessary: demonstratives, proper names, and natural-kind terms, for example, refer directly, through the agency of the speaker. Frege intentionally avoided the vexing question of why and how words refer to objects and concepts because he was weary of the intuitive answer, later alluded to by H. P. Grice, that users (minds) determine these linkages and their corresponding truth-values. Speakers use language to manipulate their listeners into believing in the manifest intentions behind their utterances. Cognitive, emotive, and descriptive meanings all emanate from speakers and their minds.
Initially, W. V. Quine put context before meaning: he not only linked meaning to experience, but also to empirically-vetted (non-introspective) world-theories. It is the context of the observed behaviors of speakers and listeners that determines what words mean, he said. Thus, Quine and others attacked Carnap's meaning postulates (logical connections as postulates governing predicates) by demonstrating that they are not necessary unless one possesses a separate account of the status of logic (i.e., the context).
Yet, this context-driven approach led to so many problems that soon Quine abandoned it and relented: translation - he conceded in his seminal tome, "Word and Object" - is indeterminate and reference is inscrutable. There are no facts when it comes to what words and sentences mean. What subjects say has no single meaning or determinately correct interpretation (when the various interpretations on offer are not equivalent and do not share the same truth value).
As the Oxford Dictionary of Philosophy summarily puts it (p. 194):
"Inscrutability (Quine later called it indeterminacy - SV) of reference (is) (t)he doctrine ... that no empirical evidence relevant to interpreting a speaker's utterances can decide among alternative and incompatible ways of assigning referents to the words used; hence there is no fact that the words have one reference or another" - even if all the interpretations are equivalent (have the same truth value).
Meaning comes before context and is not determined by it. Wittgenstein, in his later work, concurred.
Inevitably, such a solipsistic view of meaning led to an attempt to introduce a more rigorous calculus, based on concept of truth rather than on the more nebulous construct of "meaning". Both Donald Davidson and Alfred Tarski suggested that truth exists where sequences of objects satisfy parts of sentences. The meanings of sentences are their truth-conditions: the conditions under which they are true.
But, this reversion to a meaning (truth)-determined-by-context results in bizarre outcomes, bordering on tautologies: (1) every sentence has to be paired with another sentence (or even with itself!) which endows it with meaning and (2) every part of every sentence has to make a systematic semantic contribution to the sentences in which they occur.
Thus, to determine if a sentence is truthful (i.e., meaningful) one has to find another sentence that gives it meaning. Yet, how do we know that the sentence that gives it meaning is, in itself, truthful? This kind of ratiocination leads to infinite regression. And how to we measure the contribution of each part of the sentence to the sentence if we don't know the a-priori meaning of the sentence itself?! Finally, what is this "contribution" if not another name for .... meaning?!
Moreover, in generating a truth-theory based on the specific utterances of a particular speaker, one must assume that the speaker is telling the truth ("the principle of charity"). Thus, belief, language, and meaning appear to be the facets of a single phenomenon. One cannot have either of these three without the others. It, indeed, is all in the mind.
We are back to the minds of the interlocutors as the source of both context and meaning. The mind as a field of potential meanings gives rise to the various contexts in which sentences can and are proven true (i.e., meaningful). Again, meaning precedes context and, in turn, fosters it. Proponents of Epistemic or Attributor Contextualism link the propositions expressed even in knowledge sentences (X knows or doesn't know that Y) to the attributor's psychology (in this case, as the context that endows them with meaning and truth value).
On the one hand, to derive meaning in our lives, we frequently resort to social or cosmological contexts: to entities larger than ourselves and in which we can safely feel subsumed, such as God, the state, or our Earth. Religious people believe that God has a plan into which they fit and in which they are destined to play a role; nationalists believe in the permanence that nations and states afford their own transient projects and ideas (they equate permanence with worth, truth, and meaning); environmentalists implicitly regard survival as the fount of meaning that is explicitly dependent on the preservation of a diversified and functioning ecosystem (the context).
Robert Nozick posited that finite beings ("conditions") derive meaning from "larger" meaningful beings (conditions) and so ad infinitum. The buck stops with an infinite and all-encompassing being who is the source of all meaning (God).
On the other hand, Sidgwick and other philosophers pointed out that only conscious beings can appreciate life and its rewards and that, therefore, the mind (consciousness) is the ultimate fount of all values and meaning: minds make value judgments and then proceed to regard certain situations and achievements as desirable, valuable, and meaningful. Of course, this presupposes that happiness is somehow intimately connected with rendering one's life meaningful.
So, which is the ultimate contextual fount of meaning: the subject's mind or his/her (mainly social) environment?
This apparent dichotomy is false. As Richard Rorty and David Annis noted, one can't safely divorce epistemic processes, such as justification, from the social contexts in which they take place. As Sosa, Harman, and, later, John Pollock and Michael Williams remarked, social expectations determine not only the standards of what constitutes knowledge but also what is it that we know (the contents). The mind is a social construct as much as a neurological or psychological one.
To derive meaning from utterances, we need to have asymptotically perfect information about both the subject discussed and the knowledge attributor's psychology and social milieu. This is because the attributor's choice of language and ensuing justification are rooted in and responsive to both his psychology and his environment (including his personal history).
Thomas Nagel suggested that we perceive the world from a series of concentric expanding perspectives (which he divides into internal and external). The ultimate point of view is that of the Universe itself (as Sidgwick put it). Some people find it intimidating - others, exhilarating. Here, too, context, mediated by the mind, determines meaning.
To revert to our original and main theme:
Based on the discussion above, it would seem that a Creator-Designer (God) needs to have had a mind and needs to have used language in order to generate the context within which he had created. In the absence of a mind and a language, His creation would have been meaningless and, among other things, it would have lacked a clear aim or goal.
IV. Goals and Goal-orientation as Proof of Design
Throughout this discourse, it would seem that postulating the existence of a goal necessarily implies the prior forming of an intention (to realize it). A lack of intent leaves only one plausible course of action: automatism. Any action taken in the absence of a manifest intention to act is, by definition, an automatic action.
The converse is also true: automatism prescribes the existence of a sole possible mode of action, a sole possible Nature. With an automatic action, no choice is available, there are no degrees of freedom, or freedom of action. Automatic actions are, ipso facto, deterministic.
But both statements may be false. The distinction between volitional and automatic actions is not clear-cut. Surely we can conceive of a goal-oriented act behind which there is no intent of the first or second order. An intent of the second order is, for example, the intentions of the programmer as enshrined and expressed in a software application. An intent of the first order would be the intentions of the same programmer which directly lead to the composition of said software.
Consider, for instance, house pets. They engage in a variety of acts. They are goal oriented (seek food, drink, etc.). Are they possessed of a conscious, directional, volition (intent)? Many philosophers argued against such a supposition. Moreover, sometimes end-results and by-products are mistaken for goals. Is the goal of objects to fall down? Gravity is a function of the structure of space-time. When we roll a ball down a slope (which is really what gravitation is all about, according to the General Theory of Relativity) is its "goal" to come to a rest at the bottom? Evidently not.
Still, some natural processes are much less clear-cut. Natural processes are considered to be witless reactions. No intent can be attributed to them because no intelligence can be ascribed to them. This is true, but only at times.
Intelligence is hard to define. The most comprehensive approach would be to describe it as the synergetic sum of a host of processes (some conscious or mental, some not). These processes are concerned with information: its gathering, its accumulation, classification, inter-relation, association, analysis, synthesis, integration, and all other modes of processing and manipulation.
But isn't the manipulation of information what natural processes are all about? And if Nature is the sum total of all natural processes, aren't we forced to admit that Nature is (intrinsically, inherently, of itself) intelligent? The intuitive reaction to these suggestions is bound to be negative.
When we use the term "intelligence", we seem not to be concerned with just any kind of intelligence, but with intelligence that is separate from and external to what is being observed and has to be explained. If both the intelligence and the item that needs explaining are members of the same set, we tend to disregard the intelligence involved and label it as "natural" and, therefore, irrelevant.
Moreover, not everything that is created by an intelligence (however "relevant", or external) is intelligent in itself. Some products of intelligent beings are automatic and non-intelligent. On the other hand, as any Artificial Intelligence buff would confirm, automata can become intelligent, having crossed a certain quantitative or qualitative level of complexity. The weaker form of this statement is that, beyond a certain quantitative or qualitative level of complexity, it is impossible to tell the automatic from the intelligent. Is Nature automatic, is it intelligent, or on the seam between automata and intelligence?
Nature contains everything and, therefore, contains multiple intelligences. That which contains intelligence is not necessarily intelligent, unless the intelligences contained are functional determinants of the container. Quantum mechanics (rather, its Copenhagen interpretation) implies that this, precisely, is the case. Intelligent, conscious, observers determine the very existence of subatomic particles, the constituents of all matter-energy. Human (intelligent) activity determines the shape, contents and functioning of the habitat Earth. If other intelligent races populate the universe, this could be the rule, rather than the exception. Nature may, indeed, be intelligent.
Jewish mysticism believes that humans have a major role to play: to fix the results of a cosmic catastrophe, the shattering of the divine vessels through which the infinite divine light poured forth to create our finite world. If Nature is determined to a predominant extent by its contained intelligences, then it may well be teleological.
Indeed, goal-orientated behaviour (or behavior that could be explained as goal-orientated) is Nature's hallmark. The question whether automatic or intelligent mechanisms are at work really deals with an underlying issue, that of consciousness. Are these mechanisms self-aware, introspective? Is intelligence possible without such self-awareness, without the internalized understanding of what it is doing?
Kant's third and fourth dynamic antinomies deal with this apparent duality: automatism versus intelligent acts.
The third thesis relates to causation which is the result of free will as opposed to causation which is the result of the laws of nature (nomic causation). The antithesis is that freedom is an illusion and everything is pre-determined. So, the third antinomy is really about intelligence that is intrinsic to Nature (deterministic) versus intelligence that is extrinsic to it (free will).
The fourth thesis deals with a related subject: God, the ultimate intelligent creator. It states that there must exist, either as part of the world or as its cause a Necessary Being. There are compelling arguments to support both the theses and the antitheses of the antinomies.
The opposition in the antinomies is not analytic (no contradiction is involved) - it is dialectic. A method is chosen for answering a certain type of questions. That method generates another question of the same type. "The unconditioned", the final answer that logic demands is, thus, never found and endows the antinomy with its disturbing power. Both thesis and antithesis seem true.
Perhaps it is the fact that we are constrained by experience that entangles us in these intractable questions. The fact that the causation involved in free action is beyond possible experience does not mean that the idea of such a causality is meaningless.
Experience is not the best guide in other respects, as well. An effect can be caused by many causes or many causes can lead to the same effect. Analytic tools - rather than experiential ones - are called for to expose the "true" causal relations (one cause-one effect).
Experience also involves mnemic causation rather than the conventional kind. In the former, the proximate cause is composed not only of a current event but also of a past event. Richard Semon said that mnemic phenomena (such as memory) entail the postulation of engrams or intervening traces. The past cannot have a direct effect without such mediation.
Russell rejected this and did not refrain from proposing what effectively turned out to be action at a distance involving backward causation. A confession is perceived by many to annul past sins. This is the Aristotelian teleological causation. A goal generates a behaviour. A product of Nature develops as a cause of a process which ends in it (a tulip and a bulb).
Finally, the distinction between reasons and causes is not sufficiently developed to really tell apart teleological from scientific explanations. Both are relations between phenomena ordained in such a way so that other parts of the world are effected by them. If those effected parts of the world are conscious beings (not necessarily rational or free), then we have "reasons" rather than "causes".
But are reasons causal? At least, are they concerned with the causes of what is being explained? There is a myriad of answers to these questions. Even the phrase: "Are reasons causes?" may be considered to be a misleading choice of words. Mental causation is a foggy subject, to put it mildly.
Perhaps the only safe thing to say would be that causes and goals need not be confused. One is objective (and, in most cases, material), the other mental. A person can act in order to achieve some future thing but it is not a future cause that generates his actions as an effect. The immediate causes absolutely precede them. It is the past that he is influenced by, a past in which he formed a VISION of the future.
The contents of mental imagery are not subject to the laws of physics and to the asymmetry of time. The physical world and its temporal causal order are. The argument between teleologists and scientist may, all said and done, be merely semantic. Where one claims an ontological, REAL status for mental states (reasons) - one is a teleologist. Where one denies this and regards the mental as UNREAL, one is a scientist.
But, regardless of what type of arguments we adopt, physical (scientific) or metaphysical (e.g. teleological), do we need a Creator-Designer to explain the existence of the Universe? Is it parsimonious to introduce such a Supreme and Necessary Being into the calculus of the world?