Philosophy and AI Part Three: Review of “Weizenbaum’s “Computer Power…”(On Tools)

Views: 1392

AI matrix head clipart, illustration

Civilisation, as distinguished from Culture in accordance with the Kantian distinction was originally “created” because of the needs human beings have of actualising(among other powers), their potential for discourse and rationality. The family itself as a social constellation proved inadequate to meet the complex needs that humans have, needs that according to Maslow extend far beyond the physiological and safety needs that are necessary to maintain the human organism in existence. Gathering and hunting in larger nomadic groups(ca 40) also failed to meet mans self-esteem or his cognitive and aesthetic needs and man therefore chose one physical site(the village) to meet a growing constellation of needs and desires which included desires for things that were in the eyes of Socrates unnecessary and “luxurious”.

This history is the history of human psuche and is recorded in documents and preserved excavated objects that we can visit in museums. What is being exhibited in these museums are the conditions necessary for the unification of several villages with the purpose of transforming our civilisation into cultures where we can meet the growing constellation of needs and desires that emerge when living conditions are transformed. Aristotle referred to the idea of “self-sufficiency” in the phase of the transformation of villages into cities(the polis). For Aristotle our villages and cities were not artefacts or “tools” for the purposes of living but rather “organic” entities because they were created in the imaginations of psuche for the purpose of solving essentially hypothetical problems. The kind of reasoning involved in this process of civilisation is a calculative form of reasoning that involves a context of discovery that searches for appropriate means to ends: so called instrumental reasoning which is to be distinguished from the kind of reasoning about ends in themselves in contexts of explanation and justification (which is the typical form of categorical reasoning that we find in cultures.)

The history of tools obviously has a role in this complex developmental sequence which reaches back into the dawn of mans beginnings when he begins to walk upright and use tools to assist in meeting his physiological and safety needs. At this point, according to Julian Jaynes(The Origins of Consciousness in the breakdown of the bicameral mind) we are not dealing with fully conscious beings or language using beings. Jaynes describes the dawn of civilisation with reference to Mesolithic Natufian tribe who were hunters around 10,000 BC but who had settled down in 50-house settlements by 9000BC. Jaynes postulates insightfully that it was around this time that the dead were buried in ceremonial graves suggesting that this was enabled by the use of proper names for people which in turn enabled a more complex emotional and cognitive attachment to the deceased:

“Now here is a very significant change in human affairs. Instead of a nomadic tribe of about 20 hunters living in the mouths of caves we have a town with a population of at least 200 persons. It was the advent of agriculture, as attested by the abundance of sickle blades, pounders and pestles, querns and mortars, recessed in the floor of each house, for the reaping and preparation of cereals and legumes, that made such permanence and population possible.”

This was more like the villages that Aristotle may have had in mind when he talked about the organic development of civilisation into the polis, an entity of about 200,000 persons( if Athens was taken to be the typical model). The head of the town was like a God who emerged as a leader because of obvious leadership skills that related to the needs of the time. Already at this time the burial procedures for these leaders were very different to the procedures for the burial of the ordinary village dweller. By 5000 BC Jaynes notes that cities of 10000 dwellers were not uncommon around the Nile delta. The facts, then support the Aristotelian thesis of the development and organic like growth of our communities in accordance with a hierarchy of needs and wants postulated by Maslow and Freud amongst others. By the time we get to Socrates harassing people in the agora over what they think they know, the potentialities of “capable of discourse” and “rationality” have begun to be actualised and the agora is the scene of so much more than sensing- moving animals. It s the place of what Wittgenstein referred to in his later work as the hurly burly of human activity that constitutes a hylomorphic “form of life”. Language, as we can see in the Platonic dialogues has advanced into a self reflective stage where it is searching for a method and the principles(arché) behind phenomena. At this stage language is no longer used unreflectively as an instrument to accomplish everyday instrumental goals. It has become the constituter of a kind of experience many wish to use the term “consciousness” to designate. Jaynes tied consciousness and language use very closely together, focussing upon the mechanism of metaphor and analogy:

“One needs language for consciousness. We think consciousness is learned by children between two and a half and five or six years in what we call the verbal surround or verbal community. It is an aspect of learning to speak. Mental words are out there as part of the culture and part of the family. A child fits himself into these words and uses them even before he knows the meaning of them. A mother is instilling the seeds of consciousness in a two or three year old, telling the child to stop and think, asking him “What shall we do today?” or “do you remember when we did such and such or were somewhere.And all this while metaphor and analogy are hard at work.”(Jaynes, 1986)

These observations are also supported by observations related to deaf children learning a sign language and the increased complexity of behaviour that then becomes possible. Is what Jaynes doing science? He certainly thinks so, and if language is a tool, then what he is talking about would be part of the Aristotelian canon of the productive sciences. For the later Wittgenstein, language was used to play games in accordance with rules which determined our moves in these games. For Kant these rules were related to the “categories” of understanding/judgement and for him language-use was more than just a game following rules it was a tool for expressing and communicating ideas and rational processes using the categories of the understanding and the principles of noncontradiction and sufficient reason.For Wittgenstein, references to language games does not rule out the telos or purposes of these games such as :

“Giving orders and obeying them……Reporting an event, Speculating about an event, Forming and testing a hypothesis…Asking, thanking, cursing, greeting, praying”(Wittgenstein, Philosophical Investigations, P.12e)

Wittgenstein continues this reflection on the different uses of language by specifically comparing the multiplicity of the ways words are used with the multiplicity of what he calls “the tools in language”. The interesting fact about these different ways of using language is the choice of beginning with the imperative form of language which we find in ethical discussions. The method of observation involved in the language-games of reporting an event would be irrelevant to the imperative uses of language. Reporting the events of promising would not, that is, be relevant to discussing whether one ought to keep promises. The explanation/justification of promising would eventually for Kant invoke the categorical imperative which orders us to act in a certain way according to a certain principle. These forms of language, Wittgenstein continues:

“are as much a part of our natural history as walking, eating, drinking, playing.”(12e)

Weizenbaum, in a move reminiscent of many analytical Philosophers, wishes to relate our use of tools to the solipsistic mental activity of imagining which is an interesting hypothesis entertained by many philosophers in the past and can be linked to Heideggers view of Kant’s work in which reason was replaced by an idea of the transcendental imagination. He does not attempt to “reduce” the part we play mentally in this process to imagining but he does say the following:

“But tools and machines do not merely signify mans imaginativeness and its creative reach, and they are certainly not important merely as instruments for the transformation of a malleable earth: they are pregnant symbols in themselves. They symbolise the activities they enable, i.e. their own use. An oar is a tool for rowing, and it represents the skill of rowing in its whole complexity. No one who has not rowed can see an oar as truly an oar.”(P.18)

The author continues with the claim that tools then become pedagogical/cultural instruments and this transcends their purely civilisation-building instrumental uses. This transcendent use is part of another telos of language which is to symbolically recreate or represent the world they are a part of. Machines are complex tools and like tools are extensions of mans natural powers. The more complex our tools become the more our relation to nature becomes less instrumental and purposive(related to necessary human needs) and more a matter of dominating or mastering the forces of nature for the purposes of unnecessary desires(e.g. the desire to win a war). Such unnecessary needs and desires have the potential to destroy all the civilisations man has managed to create since 10,000BC in a relatively short period of time. This process began with the simple creation of vessels that could transcend the limitations of our locomotion in a space which was confined to a particular territory, e.g. ships would eventually circumnavigate the earth, continued with telescopes, which because of the intimate correlation of space with time in our space-time-matter continuum, gave the impression that we can observe the passing of time which as a matter of fact, is a mistaken description of what is going on and may well end not with a whimper but a bang on a scale difficult to comprehend.

Time, as Weizenbaum notes could be calculated by events such as the shadows cast by the movement of the sun around objects or by the time it takes us to complete various everyday essential tasks such as walking to the agora in the morning and returning during the evening or as Ecclesiastes claimed all purposes are ultimately enclosed in the life that comes to be in the world (being born)and ceases to exist (dying):

“To everything there is a season,
A time for every purpose under heaven:
A time to be born, And a time to die;
A time to plant, And a time to pluck what is planted;
A time to kill, And a time to heal;
A time to break down, And a time to build up;
A time to weep, And a time to laugh;
A time to mourn, And a time to dance;
A time to cast away stones, And a time to gather stones;
A time to embrace, And a time to refrain from embracing;
A time to gain, And a time to lose;
A time to keep, And a time to throw away;
A time to tear, And a time to sew;
A time to keep silence, And a time to speak;
A time to love, And a time to hate;
A time of war, And a time of peace.”
(NKJV)

The time of human purposes has, we know, been transcended by the scientific calculation and explorations via telescopes of the surrounding universe and this perspective, perhaps more than any other diminishes the importance of the time Ecclesiastes talks about. What Ecclesiastes discusses however, is probably typical of the kind of discussions that were being conducted in the Athenian agora in the name of eudaimonia(leading the good spirited flourishing life). The time of the scientist and the time of the prophet, i.e. are at odds with each other, requiring the philosopher to find a middle way, Aristotle’s definition of time attempted this path of the golden mean:

“Time is the measurement of motion in terms of before and after”

This definition, in the scientific view has been surpassed by Einsteins relativity theory in which it is claimed that whether an event is simultaneous with another event or comes before or after it, is relative to a system of coordinates. Weizenbaum refers to this moment as the creation of a new scientific reality which supplanted the old reality (the one defined by Aristotle?). It is not clear what is being said here. The author quotes Mumford’s claim in the work”Technics and Civilisation”:

“The clock literally created a new reality….It is important to realise that this newly created reality was and remains an impoverished version of the older one, for it rests on a rejection of those direct experiences which formed the basis for, and indeed constituted , the old reality…The rejection of direct experience was to become one of the principal characteristics of modern science.”(P.25)

The author also invokes the remark by Karl Pearson in 1892 that:

“The scientific man has above all things to strive at self elimination in his judgments”(P.25)

This flies in the face of Aristotle’s account in which the self is vital to both defining the now of his experience and in arranging the nows in a before-after continuum. If, for example, Socrates is convicted after giving his apology in court, there is no possible world in which he could have been convicted before that speech. Weizenbaum then surprisingly claims that this attempt to eliminate the self involves eventually the attempt to transform human intelligence into an artificial form of intelligence of the kind we find in our machines. That the clock is the machine that helps to bring about this, is part of the authors case as is the claim that the computer was helping humankind avoid several impending crises that would inevitably hinder the development of civilisation into a more advanced form of culture. Apparently the military were feeling the need to analyse large volumes of information and felt the machine was the most appropriate solution to this problem- an attitude that reaches back to the first of the men, Descartes who was spending his time helping the military design the war machines/tools of the time. There is no doubt that under certain time constraints the ability to calculate quickly is advantageous and for those kind of tasks the computer is indispensable. But where speed is not the decisive factor, it is not clear that computers are significantly helping us to eliminate the self from the chain of information-processing and decision-making. The denial of the self, i.e. may be a form of an attempt to master the self which requires denial the knowledge we have accumulated in relation to the self, over millennia. Insofar as this is happening it is the extent to which man is becoming more zombie-like, more like the slave-like machines he has designed to satisfy his necessary and unnecessary needs and desires.

Phenomenology responds to the challenge to reinstate experience by the curious attempt to abandon all normal forms of explanation and justification in favour of what it calls “description”. It does this via what is calls a “phenomenological reduction” which “places the world in brackets”, a process which refuses to entertain the kind of “objectivity” conceived of by scientists– a process so closely allied with the process of “observation. Heidegger is one of the foremost phenomenologists and has this to say about one important mode of “Being-in-the-World:

“The Being of those entities which we encounter as closest to us can be exhibited phenomenologically if we take as our clue our everyday Being-in-the-World……The kind of dealing which is closest to us is as we have shown, not a bare perceptual cognition, but rather that kind of concern which manipulates them and puts them to use: and this has its own kind of “knowledge”(Being and Time P.95)

Here we see the bracketing of our interpretative tendencies operating in a phenomenological reduction and the revelation of a type of concern which otherwise remains concealed. Heidegger refers to the Greek term of “pragmata” and translates this as “equipment”. Another possible term for this type of concern is “instrumentality”. The essential characteristic of such an instrumentality is, that it is a relational characteristic in which there is an assignment or reference of something to something. In such a totality of instrumentalities each item “belongs” to the others in the way in which an ink-stand, pen, paper, ink, table, lamp, furniture, windows, doors and room belong in a connected totality. The pen signifies the skill of writing and Heidegger calls this meaning of being “ready-to-hand”, a form of knowledge which manifests itself in a kind of awareness he calls circumspection–an awareness which is a typical non-observational form of awareness often encountered in all forms of acting and working. Writing is the work involved , e.g. writing an article or a book or essay. In this work we also encounter entities of the kind of Dasein(human Being-there) which have an essentially temporal nature. Time, as Ecclesiastes suggests is linked to this kind of concernful circumspection and even the clock itself is “used” as a tool. Should anything go wrong in this kind of “dealing” with the world, e.g. the clock has stopped, then the clock emerges from the totality in which it is embedded as a “thing” that can be observed “present-at-hand”. It loses its essential character as part of the instrumentality of writing at the table by the window in the study. The clock becomes “conspicuous” and loses its insertion in the totality of equipment needed for the work. When my writing is disturbed by the clock that has stopped, our circumspection is transformed into a form of consciousness of the tool which Heidegger calls “Presence-at-hand”, a form in which it emerges from its immersion in a totality of involvements into an observational field and becomes the object of a judgement “The clock has stopped”.

William James in his work “The Principles of Psychology” noted that in learning a sequence of skills that constitute a work-project, e.g. learning to play the piano, observational consciousness of the position of ones fingers in relation to the keyboard is an essential part of the skill until the learning process has been completed when the fingers and the key board form relations to each other which are essentially non-observational and sub-conscious. This kind of skill was obviously important for all civilisation-building activities. So my relation to the words I speak is not that of someone listening to what I am saying but rather is a part of this relation to Being that Heidegger calls “ready-to-hand”, my words are like tools that are used circumspectfully and are inserted in a totality of involvements that Wittgenstein urges us to analyse non-phenomenologically. Certainly when I speak, insofar as Wittgenstein is concerned, I intend to speak but one should not then attempt to describe my state of mind but rather describe human institutions and customs:

“Insofar as I intend the construction of a sentence in advance, that is made possible by the fact that I can speak the language in question”(Philosophical Investigations, 108e)

Wittgenstein also claims in this context that mastery of a language resembles mastery of a technique in which we follow firstly, the rules of language-games but secondly, also the point of the language game we are playing. The whole Phenomenological method, insofar as it aims at describing acts of consciousness or acts of thought is mistaken Wittgenstein argues, but he does admit that there are phenomenological problems which are best approached via examining the use of language. It is, Wittgenstein argues in hylomorphic spirit:

“…only of a living human being, and what resembles(behaves like) a human being can one say:it has sensations: it sees: is blind: hears: is deaf: is conscious or unconscious.”(PI 97e)

The above could well be used as a list of reasons explaining why AI does not resemble or behave like an intelligent human being. Heidegger also would share this judgement about machines and claim that a machine has no relation to Being of the form of “Being-in-the-world” or the human form of “Being-there”. He interestingly also sees the connection of instrumentalities (such as language), and signs, which he sees as a kind of universal relation. Signs in the form of equipment he argues “show or indicate”. Referring to something he claims is also a relation. Signs refer but they also have a special relation to Being, expressed in the following:

“A sign is not a Thing which stands to another thing in the relationship of indicating: it is rather an item of equipment which explicitly raises a totality of equipment into our circumspection, so that together with it the worldly character of the ready-to-hand announces itself.”(Being and Time, P.110)

Heidegger then elaborates upon this by claiming that “the Being of words and of language” are founded upon “significance”.(P.121) There is a certain affinity with Aristotle’s insistence that the power of discourse is part of what is essentially human:

“The way in which discourse gets expressed is language. Language is a totality of words—a totality in which discourse has a “worldly” being of its own…..Language can be broken up into word-Things which are present-at-hand…Discoursing or talking is the way in which we articulate “significantly” the intelligibility of Being-in-the-World…..Talking is talking about something.”(Being and Time P.204-6)

A computer “language” shares very few of these characteristics and perhaps this is as much as a metaphor as “intelligence” is when used in the phrase “Artificial intelligence”. A computer language is not used for discourse but is rather a kind of code that relates to another code:machine code.

Heidegger very rarely engages directly with Aristotle but he does often speak about the Greek language:

“The Greeks had no word for “Language”: they understood this phenomenon “in the first instance” as discourse. But because the “logos” came into their philosophical ken primarily as assertion, this was the kind of logos which they took as their clue for working out the basic structures of the forms of discourse and its components. Grammar sought its foundations in the “logic” of this logos. But this logic was based upon the ontology of the present-at-hand.”(Ibid. P.209)

This is not strictly true for the thought of Plato and Aristotle who were careful to separate the “forms” of the true and the good. Aristotle, for example, claimed in his ethics that all human activity aimed at the good, and the good was therefore one of the many meanings of Being. The logical principles of practical reasoning (noncontradiction, sufficient reason) may be the same as those operating in theoretical reason but the differences between an imperative assertion : “We ought to keep promises” and a knowledge claim, “All men are mortal”, are significant and reducing the one to the other involves ignorance of these categories and violations of the principles of noncontradiction and sufficient reason.. Heidegger clearly feels he is liberating discourse from logic, but what he in fact is doing is invoking and attacking one of the tenets of “the new men” who sought to demolish the meaningful structure of ethical discourse and other speculative forms of discourse. Liberating modern discourse of course is important work. But accusing the Greeks of ontological mistakes when two of them gave us the “tools” to avoid this mistake is a somewhat surprising move.

Leave a Reply