The Restless Clock: A History of the Centuries-Long Argument over What Makes Living Things Tick

From Clockworks2
Revision as of 17:06, 1 June 2021 by Erlichrd (talk | contribs)
(diff) ←Older revision | Current revision (diff) | Newer revision→ (diff)
Jump to navigationJump to search

Riskin, Jessica. The Restless Clock: A History of the Centuries-Long Argument over What Makes Living Things Tick. Chicago and London: U of Chicago P, 2016. "544 pages | 9 color plates, 51 halftones | 6 x 9 | © 2016."[1]


A richly illustrated, copiously annotated fulfillment of the promise of the subtitle, indexed, with over 60 pages of bibliography.

Starts with "Introduction: Huxley's Joke, or the Problem of Agency in Nature an Science." Not much of a joke, but a mild mocking of the idea that some "vitality" gives life by asking if the qualities of water come from "aquosity" (p. 1) or can eventually, in theory, be accounted for totally by the chemistry of H2O. See for the key background idea that "the core paradigm of modern science" is a mechanistic view that "describes the world as a machine — a great clock, in seventeenth- and eighteenth-century imagery — whose parts are made of inert matter, moving only when set in motion by some external force [...]": passive and without agency (p. 3).

Chapters

1. Machines in the Garden
2. Descartes among the Machines
3. The Passive Telescope or the Restless Clock
4. The First Androids
5. The Adventures of Mr. Machine
6. Dilemmas of a Self-Organizing Machine
7. Darwin between the Machines
8. The Mechanical Egg and the Intelligent Egg
9. Outside In (with sections on "The First Robots," "[...] A Nineteenth-Century Smart House," Alan Turing, and Norbert Wiener)
10. History Matters (with a highly useful brief discussion of Erwin Schrödinger's What Is Life [1944][2])

From the publisher's extended blurb: Contrary to the orthodox scientific view discouraging anthropomorphism and assigning agency to material nature and the vast majority of living beings,

[...] Since the seventeenth century, many thinkers have made agency, in various forms, central to science.

The Restless Clock examines the history of this principle, banning agency, in the life sciences. It also tells the story of dissenters embracing the opposite idea: that agency is essential to nature. The story begins with the automata of early modern Europe, as models for the new science of living things, and traces questions of science and agency through Descartes, Leibniz, Lamarck, and Darwin, among many others. Mechanist science, Jessica Riskin shows, had an associated theology: the argument from design, which found evidence for a designer in the mechanisms of nature. Rejecting such appeals to a supernatural God, the dissenters sought to naturalize agency rather than outsourcing it to a “divine engineer.” Their model cast living things not as passive but as active, self-making machines.

The conflict between passive- and active-mechanist approaches maintains a subterranean life in current science, shaping debates in fields such as evolutionary biology, cognitive science, and artificial intelligence. [...][3]

++++++++++++++++++++++++++++++ ++++++++++++++++++++++++++++++

Chapter 1, "Machines in the Garden" (note Leo Marx's, The Machine in the Garden)

Relevant here primarily for its extended discussion and illustration of of late Medieval and Renaissance (or "Early Modern") automatons of various sorts: complex, ingenious clocks and miniature mechanical theaters, hydraulic engines, etc. for a wide group. Note well that along with these popular devices could go a denial of a division between mechanical and living in that the more sophisticated automata could be experienced as "lively" in more than a figurative sense.

++++++++++++++++++++++++++++++ Chapter 2, "Descartes among the Machines" (note Samuel Butler's pseudonymous "Darwin Among the Machines")[4]

Descartes's proposal that an animal is a machine has sounded to most people [...] like saying that an animal is essentially inanimate. [...] But it is a misreading of Descartes: the key point about his animal machine was that it was alive; it was a living machine. [...] To be alive was the whole purpose of Descartes's animal-machine. Not as if alive, not apparently alive [like automata discussed earlier], but actually alive. By describing animals as automata, Descartes did not mean to reduce them to lifelessness. On the contrary, he meant to declare that one could explain every aspect of life in terms of machines, and so could understand the workings of living beings as fully as a clockmaker understands a clock. Rather than to reduce life to mechanism, he meant to elevate mechanism to life [...]. (pp. 44-45)

We will note here that this reading of Descartes (rightly or wrongly) led to some mocking his theory of the soul in a human body as "the ghost in the machine," which has become a catchphrase.[5]

Chapter discusses not only Descartes but key figures who preceded him, and then contemporaries and successors who read, misread (?), supported, and argued with him, including such notables in the mechanism debate as Aristotle and the theory of three souls (vegetative, sensitive, rational),[6] the ancient physician Galen, John Locke, the Dutch Catholic theologian Caterus, who "warned that Descartes had note offered no good defense against those who believed that the human mind was a bodily thing" (p. 65), Thomas Hobbes, and later Georges Buffon and Voltaire.

++++++++++++++++++++++++++++++ Chapter 3, "The Passive Telescope or the Restless Clock"

"The telescope [...] and the clock both served as key models of living structures" — eyes and optics, especially for the telescope — "in the new mechanist life sciences of the mid- to late-seventeenth century" (p. 77). Of interest here is the clock analogy, leading on one side to a vision of "the new animal-machine drained of life," as William Coward had us: "a meer [sic] piece of Mechanism," if (amazingly) also "a Reasoning engine made out of dead matter" (quoted, Riskin pp. 78-79). Mechanism, however, was complex, and very much so in the thought of Gottfried Wilhelm Leibniz. Like everyone at the time writing on such matters,

[...] Liebniz described animal and human bodies as equivalent to automata. But he meant something very different by it from what was rapidly becoming the dominant meaning. The shared keywords of the seventeenth century — "mechanical," "clockwork" — camouflaged radical differences of opinion. In Leibniz's view, neither animals nor, indeed, machines were passive or brute. Leibniz was among the many who rejected Descartes's claim that animals lacked souls. But in his case it was part of a more general refutation of Cartesian physics: to Leibniz, nothing really lacked a soul. (p. 95)

For this view "Artificial mechanisms provided models not only for agency, but also for indeterminate, variable[,] and responsive activity" (p. 95). "'All of nature,' Leibniz wrote, 'is full of life,' and exactly as full of mechanism" (Riskin p. 108).

++++++++++++++++++++++++++++++ Chapter 4, "The First Androids"

These "androids" are not the mostly organic entities of current usage, but — as we should expect from Riskin's thesis — neither are they robots. The first on record goes back to 1677, the creation of Salomon Reisel, a physician and "philosophical mechanist" who claimed to have built an "'artificial man' with all the internal bodily functions," and with Reisel planning "to endow his creation with speech and the ability to move about on its own." Described in a journal article of the time as something we might see like a 1960s museum model of Our Amazing Body "'with such similarity & resemblance to man in all internal parts that, except for the operations of the rational soul, one sees in it all that happens in our bodies, & this by the principles of Physics-Hydrostatics'" (Riskin p. 113). "The first androids, then, seemed anything but clocklike in the traditional sense of the term; they seemed clocklike, rather in Leibniz's sense: unquiet, restless, visceral, active, responsive" (p. 132).

Note Wolfgang von Kempelen's hoax of the "Chess-Playing Turk" (continued by Johann Maelzel), which, in its long career, (a) defeated among others Benjamin Franklin and Charles Babbage, and (b) somewhat ironically was a serious contribution to the debate "whether intelligent mental processes could be reproduced by artificial machinery" (pp. 123-25 f.), and relevant for discussions of Descartes and "the Cartesian divide" (p. 125), and AI. Also see this chapter for

A mechanical duck that shat, or appeared to, and was sufficiently famous to appear (inaccurately drawn) in Scientific American for 21 Jan. 1899 (Riskin pp. 133 f.): for at least a simulacrum of the superimposition of the mechanical on the quite basically organic.
A fad in the 1730-40s of "moving anatomies," machines that "breathed and bled," or appeared to (pp. 136 f.) — with proposals for even more completely lifelike automata, some suitable for medical experimentation. 
With the more ambitious, there was the continuing desire to make an automaton that would simulate all human functions, including "apparently crossing" that "Cartesian boundary between mechanical body and rational soul" by producing "'speech and the articulation of words'" (Riskin p. 137, quoting a source from 1744). So in "the last decades of the eighteenth century," there was a time "when philosophers and mechanicians and paying audiences were briefly preoccupied with the idea that articulate language was a bodily function" and hence "that Descartes's divide between mind and body might be bridged in the organs of speech" (p. 144).

In the mid-18th c., "Automata were a preoccupation among [... a number of] modernizing proprietors, imperialists, and slaveholders. The equivalence of machines to lowly people of various sorts — slaves, conscripts, workers — was an attractive supposition and a theme that would recur throughout the development of automatic machinery, industrial as well as experimental. In such conversations, machines unsurprisingly had a decidedly Cartesian meaning: they signified the lack of a rational soul, of a capacity for reason and intellect. (p. 146)
Mechanical calculators had an analogous effect to the automatic loom demoting calculation from a paradigm of intelligence to the antithesis of intelligence. If a  machine could calculate, then something else — say, decision making or language — must be emblematic of human intelligence.

— Participating in this discussion: such big names as Blaise Pascal, Leibniz, and Charles Babbage (a key worker in fields leading to computers (p. 148 f.).

Riskin notes changes in the idea of "machine" in and a bit beyond 18th c., with the five initial editions of the Dictionnaire of the French Academy defining the term, "'Engine, instrument sutured to move, pull, lift, drag or throw something,' with figurative uses such as 'man is an admirable machine.'" In the 6th edition of 1832-35 the example for one figurative use is "It is nothing but a machine [...] a walking machine, [for] a person without spirit, without energy" (p. 150).

A brute-mechanistic distinction between mind and mechanism informed the process of industrialization and that process made manifest an ultimately brute-mechanistic division of the social and economic world into parts, mind on one side and mechanism on the other. The new class of automaton workers included humans and animals as well as machines. The industrial reformers and inventors of automated machinery alike understood their task as the ingenious and lucrative division of intelligence from labor, design from execution, agency from mechanism. (Riskin p. 150)

++++++++++++++++++++++++++++++

Chapter 5, "The Adventures of Mr. Machine" Opening section on Julien Offray de La Mettrie, author most influentially of Man a Machine (French: L'homme Machine), 1747, which see. Riskin discusses insightfully and in some detail how the human-machine has agency, thought, and the capacity for a sex life (pp. 153 f.).

Later parts of the chapter — "The Sentimental Education of Mr. Machine"[7]— also deal largely with French philosophy — in a kind of Who Was Who of 18th-c. philosophical writing — and very important on how the human-machine idea could be used for materialist arguments for "natural" hierarchies in terms of race, class, gender, and to some extent nationality: some machines are clearly more sensitive, efficient, and elegant than others (and so should rule). Alternatively, one could support the mechanistic model and still argue for human equality. Having this discussion in the background might usefully highlight the fabular aspects on race and slavery in such works as Isaac Asimov's "The Bicentennial Man," and the TV episode Star Trek: The Next Generation, "Data's Day".

Riskin concludes this section on the argument — with some philosophers taking both sides at one time or another — with a sardonic summation of what she has just illustrated through extensive quotations from these thinkers themselves (and she does not approve):

We are all just stuff. But the world of stuff is infinitely divisible. The human-machine model and the conversation about the relations between human-machinery and human agency underlay a rapid alternation between universalist ideals on the one hand and the whole panoply of relegations on the other: Hottentots are mechanically indistinguishable from orangutans; female machines are driven by the womb; faulty machines such as the deaf and blind are not really human and so on. This dizzying alternation was itself a hallmark of the [contentious and contradictory] Enlightenment discussion of human nature. We are all the same, a single type of living machine. [Vs.] We are arranged on a graduated scale of mechanical perfection from [the philosopher] Montesquieu down to a monkey, from [inventor, artisan, and artist Jacques de] Vaucanson to his automata. (p. 178)

Section, "Organized Rather Than Designed, Mr. Machine Evolves"

Mr. Machine, that moral material creature, was no passive, brute mechanism. On the contrary, he was an active, self-moving, self-constituting mechanism. No divine Clockmaker assigned him his structure, function, or source of movement. These developed from within himself. [...] Mr. Machine was not a designed machine but an organized one. (Riskin p. 178)

We will note that, like Koheleth (3rd c. BCE?), La Mettrie, Buffon, et al. among anti-Rationalist French philosophers declined "to set humans, and especially themselves, apart from the rest of nature with the idea of a disembodied rational soul [... insisting we] must learn the 'humiliating truth' that man was an animal" (p. 181).[8] These ideas can serve for deep background on the slow-motion debate in 20th-c. SF on whether the embodiment of an uploaded/downloaded personality affected the personality: for primary examples, "No Woman Born," "Masks," and, in Frederik Pohl's Gateway series, The Boy Who Would Live Forever. We will add that the idea of organization as central to life proved robust, and that these Enlightenment ideas prefigured later theories of emergent properties.[9] They also look backward and forward to theories and narratives — in a way that can feel paradoxical — that link materialist humanity to the larger world: "A living, sentient cosmos" (cf. and contrast "the Gaia hypothesis" for Earth) "thrumming with feeling traveled arm in arm with Mr. Machine, a further effect of eliminating Descartes's disjunction between mechanism and self, self and world" (Riskin p. 184).

++++++++++++++++++++++++++++++

Chapter 6, "Dilemmas of a Self-Organizing Machine" And following the Enlightenment (and the "Augustan Age" in English — RDE), Romantics:

The man-machine's quandary was a chief preoccupation of the Romantic movement,[10] as this period is traditionally known in the history of literature and the arts, and more recently, in the history of science too. The Romantics struggled mightily with the idea [...] that living beings might be self-organizing and self-transforming machines, striving to constitute and reconstitute themselves in the dynamic, living machinery of nature. [...]

Section, "Striving Machines" The condition of a self-organizing living machine inhabiting an un-designed world inspired an effusion of poetry and science — or to put it better, an effusive fusion of poetry and science — around the turn of the eighteenth to the nineteenth centuries. A remarkable intimacy between poetry and science characterized the Romantic movement. (p. 189)

And the Mechanism vs. Vitalism debate was proceeding robustly in what was developing in the period into truly scientific biology.


Deals briefly with Jean-Baptiste Lamarck, but notes his centrality to the succeeding chapters. "When he coined the term biologie in 1802, Lamarck defined the [...] new field in terms of [...] a vital, mechanical striving. An intrinsic pouvoir de la vie (force of life) [...] drove 'animated machines,' plants and animals, not only to compose themselves, but to elaborate and complicate the organization over time" — a very long time (p. 199). "A living thing was like a watch," for one of the comparisons standard by Lamarck's time, but "only insofar as a watch could be considered to contain the agency that set it in motion," implying the robust idea of "life as a form of activity," and "a living being" as "an agent" (p. 201). {Cf. and contrast more recent debate between orthodox behaviorists[11] with, so to speak, the slogan "Animal non agit, agitur" (animals do not act; they are acted upon [sic on plural for English] vs. the ethnologists' idea that healthy animals are "up and doing."[12]

Section, "The Striving Machinery of Life Dramatized" Samuel Taylor Coleridge (key English Romantic): "The dilemma that Kant had struggled to describe and overcome was the focus of Coleridge's writing. Rather than working to resolve it, he dramatized it. Descartes's mechanist system, Coleridge wrote, was a 'lifeless Machine whirled about by the dust of its own Grinding,' a reduction of "the living fountain of Life' to 'Death.' On the contrary, in Coleridge's judgment, living organs were different from artificial machines in that, rather than being composed of parts, they actively assimilated foreign matter into themselves" by an "'unseen Agency,'" with Riskin seeing "Agency" as "the key word in Coleridge's understand of living nature" (p. 202).

Moves on to Frankenstein (with the Creature "the leading hypothetical man-machine of the Romantic period"): with background in "machinery as a model of animal and human life" and (renewed) theories of spontaneous generation of life (203-04). "Frankenstein's monster represented the central dilemma of contemporary science" when the book was written, "according to which all living beings were constituted by an inherent agency and yet made out of dead matter." More generally, "Rather than contrasting life with nonlife — the inanimate — the Romantics set life up against death. What was not alive was dead" (p. 207). "That matter was dead, the opposite of life, and that life was a form of activity[...] were the Romantic principles that informed the founding of biology as a discipline. Life was the struggle against extinction [...]" (p. 208). Note rising popularity of ideas on "Vital Power" and "the World Spirit, and the entry into the a-borning new science of physiology as well as biology generally Leibniz's idea of vis viva ("living force") and how, with this concept, Leibniz "had built the source of action into his world-machine rather than attributing it to an external source, distinguishing his intrinsically active form of mechanism from what he saw as the passive machinery of both Descartes's and Newton's cosmoses" (pp. 209-10).

Chapter ends with consideration that "The dead matter of the Romantics became animate, not at the hands of an eternal Designer, but through the action of a vital agency, [...] an all-embracing energy intrinsic to nature's machinery" (p. 212). Which lead to two developments: (1) the idea "that by tracing the developments of limited agents working in specific contexts over [long] periods of time, they might reconcile the demands of mechanism with the appearance of living purpose. [...] [T]ranscendence through the dimension of time." And (2) "a kind of transcendence though energy: the idea that organisms, understood as living machines, were connected through a great web of energy exchange to the cosmos itself" — a web of energy that included Frankenstein's "Monster" (pp. 212-13)


See chapter for the long-running debate on Mechanism vs. Vitalism; see also for similar works from the period, such as E. T. A. Hoffman's "Automata."

++++++++++++++++++++++++++++++

Chapter 7, "Darwin between the Machines" Notes Romanticism's effects on Charles Darwin, however much he was no fan of his grandfather's, Erasmus Darwin's, exercises in scientific Romantic poetry, and directly through the influence of the German Romantic Alexander von Humboldt (p. 215).

It was a deep, an elemental ambivalence [... in Darwin's ideas of mechanism]. Moreover, Darwin was drawn to the work of other people torn between the same two impulses: to banish agency from living nature's machinery and to make agency its very crux. The French physiologist Claude Bernard struggled mightily with the same problem, also without solving it, and also extremely productively. [...] ¶ "An organism is nothing but a living machine," was Bernards constant refrain. [...] Moreover, living machines did not differ from "brute machines" [...] ¶ Yet at the same time, Bernard did also believe that living machines had a certain distinguishing feature, unique to them [...]: "a creating idea [...]". The vital idea, Bernard said, was the essence of life. [...] Indeed, he continually used the term "vital force" all the while insisting that there was no such thing. [...] Bernard rejected both what he called "vitalism" [...] and mechanistic materialism. (pp. 235-36)

Darwin's supporter T. H. "Huxley argued that Descartes's idea of living beings as machines should be extended to include even the human mind: 'We are conscious automata'" (p. 237). Darwin remained ambivalent: His "mechanist accounts of life [...] encompassed both extremes: a kind of machinery in which agency was elemental, constitutive of the world, and a kind of machinery in which agency was illusory, reducible to brute parts" (p. 238).

No agency, though, was an invitation to accept William Paley's 1802 thought-experiment where one finds a watch on a heath and infers "A watch implies a watchmaker: it is a perfect encapsulation of the argument [for God] from design. What people do not generally notice is the part played by the watch here: it lies upon the heath like a stone about to be kicked. The watch is perhaps unlike a stone in its mechanical complexity but in Paley's famous passage it resembles the stone in its passivity and inertness [...]. It represents the clockwork universe of early modern science not only in being an intricate mechanism, but also in being a profoundly passive object of externally imposed design" (p. 239). This became more of a problem for Darwin when he accepted the term of "fitness," since this idea "retained a ghostly aura of the divine Engineer. Through they had evacuated God from his machinery, mechanists and authors of argument from design had left behind a device world, an artifact world: a world utterly dependent upon an external source of purpose and action" (p. 243).

+++++++++++++++++++++++++++++ Chapter 8, "The Mechanical Egg and the Intelligent Egg" Covers the the beginnings in Europe of embryology as an independent scientific discipline; but introductory section offers some significant general background.

[... In the growing consensus] Darwinism was unequivocally "mechanist," notwithstanding Darin's own ambivalence about the mechanical models of life [...]. Darwinism became "mechanist," and "mechanist," in turn, became the opposite of "vitalist," a word often a[[;ed to science of the eighteenth century and even earlier. but which was [...] in fact an innovation of the nineteenth. [... Our emphasis]

Yet a tradition [...] going back to the earliest ones to describe species as changing in time [...] understood their approach as at once mechanist and historical. They took living entities to be machines, by which they meant rationally comprehensible systems of moving parts that developed and changed thanks to a kind of historical agency: an ability throughout the parts of the machinery to constitute and transform themselves and their relations to one another purposefully and responsively. (p. 252) * * *

[Max] Weber admonished that to be successful, one must accept that science was not a path to meaning, art, God, or happiness. Science dealt neither in the revelations of "seers and prophets," not in the truths oaf "sages and philosophers." What science had to offer was technology, the means to carol the conditions of daily life, and also clear thinking." A key role of this German Protestant mode of positivist, mechanist science was to puncture the authority of generalized visions. (p. 256)

Sections follow on "The Mechanical Egg and the Historical Egg" — "eggs and embryos as machines and, specifically, not [...] developing historical entities" (p. 257) — for one view; and "The Homunculus Machine and the Self-Made Embryo" for two competing views on ontogeny and its relation to phylogeny (etc.), plus analysis, not immediately relevant here, of August Weismann's important contributions in starting the road to genetics and, more relevant but less admirable, Weissmann's caricaturing Lamarck's views on "the inheritance of acquired characteristics," a formulation Lamarck never used (p. 268). Moving into the 20th c., Riskin's quotes a 1916 text, she finds characteristic:

We tend to lay the causes of form-change, of evolution, as far as possible outside the living organism. With Darwin we seek the transforming factors in the environment rather than within  the organism itself. We fight shy of the Lamarckian conception that the living thing obscurely works out its own salvation by blind and instinctive effort. We like to think of organisms as machines, as passive inventions gradually perfected from generation to generation by external agency, by enviornmnet or by natural selection or what you will" (quoted by Riskin p. 272)

Quotes Weismann disciple Erwin Stresemann with use of a watch-word with later (psychological) Behaviorists, if not always with the "sed": "animal non agit, sed agitur" — an animal doesn't act, but is acted upon — part of a larger "Weismannism" that would "dominate biological thinking in the United States in the early-to-middle decades ion the twentieth century" (Riskin p. 278) — challenged in the 1960s and after by, among other views, the idea of ethologists that healthy animals are "up and doing" (Konrad Lorenz, as remembered by RDE).

Section, "The Intelligent Egg and the Dead Sparrow"

Hans Driesch of the sea urchin egg experiments:[13]

Thus if the embryo were a machine, it must be a machine of which any given part [...] could do the job of the whole. "A very strange sort of machine, indeed, which is the same of all its parts!" In the face of these results [whole — if smaller — embryos developing even from a single early-embryonic cell], Driesch wrote, "the machine theory as an embryological theory becomes an absurdity. These facts contradict the concept of a machine; for a machine is a specific arrangement of parts, and it does not remain what it was if you remove from it any portion you like." (Riskin p. 288) 

Notes Jacques Loeb's 1912 manifesto, The Mechanistic Conception of Life, "which made the reductive elimination of agency from living creatures into both the condition and supreme goal of mechanistic science," which Loeb (following Positivist Ernst Mach) championed (pp. 289, 291), and which made him, if he bragged truthfully, condemned by the arch-Vitalist Henri Bergson "one of the most dangerous men in America" — and, Riskin says, the model for Max Gottlieb in Sinclair Lewis's Arrowsmith[14] (p. 290).

As with earlier human-machine manifestos, Loeb's presented an overtly ethical theory. The fact that human beings were "chemical mechanisms" [...] must inform a modern, scientific ethic at the root of which lay the instincts: "We eat, drink, reproduce not because mankind has reached agreement that is is desirable, but because, machine-like, we are compelled to do so." The same must be true of the very highest and noblest of impulses, human beings' willingness "to sacrifice their lives for an idea." These ideas might give rise to "chemical changes" in the body that made people "slaves to certain stimuli just as the copepods [small crustaceans (Riskin's addition)] become slaves to the light when carbon dioxide is added to water."

"Slaves" with no "freedom of motion": this was Loeb's ethical-scientific vision of living creatures, including human beings. [...]

For Loeb, recognizing the mechanical enslavement of all living beings was important not only scientifically but morally and politically, as an inoculation against the Romantic, metaphysical mythologies by which human beings enslaved one another. World War I persuaded him that a brute-mechanist theory of life was essential for a Germany that had fallen into the metaphysical and Romantic tendencies of its aristocratic Junker class. (pp. 292-93)

Riskin has Driesch and Loeb personifying "the radicalization of their respective positions around the turn of the twentieth century." On the one hand, "an inexorable vital force" (Driesch) and on the other a mechanism that had Loeb seeing a sparrow flying down to get a seed as having no more volition than a dead sparrow dropping on the seed. "On neither side does one find the active mechanism of old: natural mechanism as a rational system of parts operating over time" (historically) "in a limited and contingent, yet intrinsically active way" (pp. 293-94).


+++++++++++++++++++++++++++++++++

Chapter 9, "Outside In"

After a brief introduction noting cybernetics[15]

Section, "The First Robots" See for a discussion of R. U. R. and fitting it into the mechanist/vitalist controversy, and developing concerns over treatment of workers (pp. 296-300), a later, more familiar (more mechanical) robots of the first third of the 20th century.

Section, "Flashback: A Nineteenth-Century Smart House": That of the French magician Jean-Eugène Robert-Houdin, 1850s-60s, filled with "electromechanical machines" (p. 205). More generally for the period —

These first responsive, sentient [sic] electromechanical machines were born [sic] just a few decades after [Charles] Babbage's projects in mechanical computing, [1820s] [16] and a good half century before the first "robots." The idea that human functions, especially the functions of socially inferior humans, might be carried out by artificial machinery, originated [...] in the early stages of the Industrial Revolution, and therefore long predated cybernetics. So too did the idea that artificial mechanical servants could be sentient and responsive: could be endowed with a kind of mechanical agency. Indeed, ,if Robert-Houdin's career is any indication, artificial, mechanical agency was a guiding preoccupation of nineteenth-century technical innovation. (p. 308)

See for K. Vonnegut's Player Piano, Fritz Lang's METROPOLIS, and related works looking at labor in The Machine Age.


Section, "Agency in the Passive Voice"

That's passive voice, deleted agent: Riskin has organisms active in achieving homeostasis, and the study of organic homeostasis central for the development of what Norbert Wiener was to call in Cybernetics: or, Control and Communication in the Animal and the Machine, cybernetics (pp. 308-09).

Why belabor Wiener's upside-down readings of history? He was a mathematician and philosopher, after all, not a historian. Yet he founded his field of cybernetics upon a vision of history. An axiom of cybernetics was that until then, no one from Hero of Alexandria onward [...] had considered the possibility that responsiveness might be mechanical: that animals and artificial machines might share mechanisms of sentience and response. Cuyberneticists maintained there new machines were the first to transcend the limits of the passive, scripted, "blind, deaf and dumb" automata of old. Although they did make an exception for Watt's self-regulating steam engine and the other governors in [James Clerk] Maxwell's 1868 paper,[17] these, they thought, had gone unrecognized as sentient and responsive machines and had thus failed to start a revolution in philosophy and engineering.

Cybernetics was this deferred revolution coming at last. (p. 312)

And that coming of cybernetics brought with it "the new generation of 'sensitive automata'" that could sense and interact with their environment. Riskin quickly reviews the history of "Responsive automata" going back to the 13th c., and from there back to antiquity (p. 312). "Not knowing history, the cyberneticists were condemned to repeat it. As a result, their mechanist modes of sentience, responsiveness, and living agency became just the opposite: so many confirmations of the absence of agency in biological creatures" (p. 313).

Such denial of agency should sound familiar to anyone who studied psychology in the middle of the 20th c.: "Removing agency, along with all subjective experience from psychology was one of the core principles of the behaviorist school then prevalent [...] The Harvard behaviorist B. F. Skinner wrote that the study of reflex action had undermined the "prevailing theories of the inner agents responsible for behavior." Going along with Descartes on animals as machines, "Skinner and other behavior psychologists [...] described an animal's ability to respond to its environment as 'machine-like' and presented their behaviors approach as scientific in its construal of creatures as machines" (p. 315). Stimulus/Response, alliance with "The cyberneticists' central project" of "the mechanization of perceptive agency" (p. 316): Animal non agit, agitur all the way.


Section, "The Passive-Active Cybernetic Menagerie": Maze-solving and similar 20th-c. machines that sensed parts of, and interacted with, their environments (316-29).

Section, "Outside In": Russ Ashby, Alan Turing, et al. and their "approaches to building a thinking machine," and getting computers (quote from p. 330). In this post-WW2 work Turing felt that

Instead of trying to build an adult intelligence directly [...] it made better sense to try to simulate a child's mind [,...with] what he called an "unorganized machine," that is, a bunch of neuron-like components connected "in an unsystematic way" [... and with] their configuration subject to some random variation. He thus laid out a blue print fro an approach to artificial intelligence that a decade and a half later would be called "connectionism." The machine would become organized by means of an education [...]. (pp. 330-31)

Thus, the first principle Turing shared with Ashby and other cyberneticists, notably Wiener, was the conviction that life and mind were crucially diachronic mechanical phenomena, developing over time. [...] Second, however, Turing [...] reached for a form of development in time, of learning, that required no actual self, no inner agent. He too drew upon behaviorist principles to describe how learning might be directed from the outside in.

In order to learn, a machine must be fallible: embracing fallibility would crash the problem of machine intelligence. (p. 331)

For maturing machines, see The Adolescence of P-1 and the brief childhood (and second childhood) of HAL 9000 in A. C. Clarke's 2001: A Space Odyssey and its successors.

Riskin asserts that for Turing "some details of its internal workings must remain unknown" for a machine "to seem intelligent"; hence Turing "carried behaviorism a step beyond the behaviorists [...]" (p. 334). The section title is explained in an important paragraph on "Turing's landmark essay 'Computing Machinery and Intelligence' (1950)," where Riskin notes "the crucial point" on the "necessity of measuring intelligence from the outside and not inside" (pp. 334-35).

Turing specified that a machine must be admissible for the test even if its engineers could not fully describe how it worked "because they have applied a method which is largely experimental." He imaged the successful machine as a "child" machine that would become intelligent by being taught, and he emphasized that even the "teacher" of the machine would "often be largely ignorant of quite what is going on inside." [...] This was necessarily the case, since to be able to describe the machine's inner workings completely would be to lose the sense that it was intelligent. But, then, the same would be true in the case of a man: if one could attribute his behavior entirely to a set of explicit rules, Turing argued, one would come to see him as a machine and, by implication, without intelligence.

Turing believed that such a set of rules existed. [...] Humans did not regulate their lives by rules of conduct. Instead they were regulated by laws of behavior: they were the objects, not the agents of regulation. (p. 335)

On raising an AI computer, see again HAL in Clarke's novel (linked above) and SAL 9000 in Clarke's 2010: Odyssey Two.

Summing up the chapter:

Several current fields of research emerged from these mid-twentieth-century projects: cognitive science, artificial intelligence [AI], robotics, artificial life [AL]. Historians and practitioners of those fields generally describe their history as shaped by a struggle between the two competing approaches. The first, "cognitivism," or "computationalism," construed thought as information processing [...]. The second approach, still according to the consensus view, is "embodied [...., not construing] thought as abstract reasoning, but as a form of physical engagement with the world. [...]

But this dichotomy hides deeper commonalities. Turing's disembodied, cognitive program included his "P-machine," trainable through "pleasure and pain," on the principle that an intelligent machine must learn from experience. [...] The "cognitivist" and "embodied" programs shared the axiom that intelligence must be an epiphenomenon of pleasure and pain, which were in turn mechanisms by which the environment directed the mechanical creature. Fundamentally, they agreed upon the essential passivity of the living, intelligent being. [...]

In their physical experiments, thought experiments, and interpretations, they understood agency in behaviorist terms, as a secondary appearance rather than a primary reality. They flirted with the intelligent egg but they married the dead sparrow. (pp. 335-36)


+++++++++++++++++++++++++++++++++

Chapter 10, "History Matters"

Bringing the story up to Richard Dawkins's The Selfish Gene (1976), Erwin Schrödinger's What Is Life (1944), and the emergence of "the notion of 'emergence'" as "an organizing idea that has permeated the natural and social sciences during the last three decades" (p. 366) — and to a polemical conclusion on why "History Matters." Significantly missing is a return to the case of Trofim Lysenko,[18][19][20] who was discussed briefly on two pages in Chapter 9 (pp. 273-74) and — on reasonable grounds — dismissed. Riskin concludes the chapter and her argument

Three and a half centuries after the natural sciences assumed their modern forms [...,] their relations to religious beliefs and institutions remain fiercely embroiled. This book has tried to show that their embroilment was built into modern science from its inception, built into a definition of science that made a supernatural God necessary to naturalist explanation by banishing agency from nature's mechanism. Practitioners of the sciences have regularly, from the beginning, challenged this brute-mechanist definition of science and violated its ban on agency. Yet they have not succeeded in dethroning the brute mechanist ideal of science, whose supernaturalism is deeply implicitly despite being long hidden. To say that a human being works like a machine, whether one accepts or rejects the idea, sounds like science. But it sounds less like science when one describes the machinery as restless, moved by its own inner agency. Historical analysis, by explaining why the first sounds are like science and the second less, can help to reopen scientific possibilities that the classical clockwork model of nature foreclosed, but that tick on within the model of a living being as a restless clock. (p. 74)


++++++++++++++++++++++++++++++

++++++++++++++++++++++++++++++

See also Sublime Dreams of Living Machines: The Automaton in the European Imagination and "The empty brain". Note The Restless Clock as potentially useful deep background for Gaiman's American Gods.



RDE, finishing, 28Feb21f. through May 2021