The Vapour Engine

From Fabulous Loop de Loop
Jump to navigation Jump to search

Samuel Butler: “We shall never get straight till we leave off trying to separate mind and matter. Mind is not a thing or, if it be, we know nothing about it; it is a function of matter. Matter is not a thing or, if it be, we know nothing about it; it is a function of mind.” [1]

Gregory Bateson: Up until Lamarck mind was an explanation of the biological world. But, Hey presto, the question now arose: is the biological world the explanation of mind?" [2]


In this chapter I will discuss the work of Samuel Butler, who made a series of claims in relation to consciousness and purpose, and in relation to machine life and organic life which were avant cybernetics.[3]

In Evolution Old and New (1879) Butler articulated a dialectical relationship between the discourse of the machine and the discourse of evolution. These claims would later be reformulated in more precisely cybernetic terms by Gregory Bateson, Norbert Wiener, Kenneth Craik, Warren McCulloch, and others.

Samuel Butler had been a strong adherent of Darwinian evolutionary theory since 1859, when he first encountered Charles Darwin’s The Origin of The Species. The two men conducted an amicable correspondence as Butler published a series of texts which took the theory of natural selection as their starting premise, including Darwin Among the Machines (1863) which speculated on the conditions necessary for a machine to evolve in the manner of an organism (a theme revisited in Butler's novel Erewhon a decade later).

In 1863 Samuel Butler sent a letter to the editor of The Press, Christchurch, New Zealand which was published under the heading, Darwin Among the Machines (and signed under the pseudonym Cellarius).[4] in which he speculated that via natural selection, machines might evolve consciousness.

“We refer to the question: What sort of creature man’s next successor in the supremacy of the earth is likely to be.” It will be, answered Butler the machines man designs, regulated by man-made contrivances which give them a “self-regulating, self-acting power”. Machines with the ability to self-regulate were already amongst us, these were the “vapor engines”, or steam engines, regulated by a governor.[5]

Erewhon with illustration by Robert Gibbings

The satirical thread running through Darwin Among the Machines is that humans should resist the rise of the machines despite the fact that these advanced machines will be in every respect better than humans. They will possess no human (all too human) imperfections such as avarice, ambition, jealousy, or impure desire. Butler muses on ways in which the general lot of humans might actually improve, as they assume their role as the domesticated life-stock of superior mechanical life-forms. As slaves who attend to their needs, humans will be treated well. The master-machines will treat humans in the same way any enlightened farmer would recognise the value of good husbandry. As machines grow in complexity and subtlety, demanding more from human lives, there will even come a point when they will self-produce. “There is nothing that our infatuated race would like more than to see the fertile union between two steam engines” writes Butler, isolating the death-wish at the heart of this very human fantasy.

Here, Butler registers a vital shift in the discourse of the machine. It is at the point that the “vapour engine” arrives on the scene that an equivalence between organic life and machines gains a new plausibility. In 1858 Alfred Russel Wallis had noted that natural selection was akin to the self-regulation of a steam engine. “The action of this principle [natural selection] is exactly like that of the centrifugal governor of the steam engine,” wrote Wallis “which checks and corrects any irregularities almost before they become evident; and in like manner no unbalanced deficiency in the animal kingdom can ever reach any conspicuous magnitude, because it would make itself felt at the very first step, by rendering existence difficult and extinction almost sure soon to follow.” [6] Wallis’ description of a self-regulating feedback loop represents another significant shift toward the discourse of the machine – which Butler will be swift to pick up on in his novel Erewhon (1872). This new relation to the machine recognises that the essential boundaries between organism and machine, organism and environment, organ and tool, conscious purpose and involuntary action are increasingly hard to maintain. As far as Darwin Among the Machines is concerned, the fictional author of Butler’s letter advocates a revolution. Humans must rebel and destroy the machines before we sleepwalk into dependency and, as the master-slave dialectic is complete, reach a point of “absolute acquiescence in our bondage”.


The possibility of mechanical evolution is developed further in Butler’s novel Erewhon: Over the Ridge (1872). The novel takes a similar form to Swift’s Gulliver’s Travels (1726), Voltaire’s Candide (1759) or William Morris’ News From Nowhere (1908), wherein the hero-traveller (in this case Higgs) encounters an unfamiliar world in which a series of topsy-turvy values are demonstrated. In Erewhon sickness is regarded as a crime and punished with imprisonment. By contrast, crime is regarded as an illness and those afflicted are pitied and rehabilitated. The young are educated in Colleges of Unreason, where they are taught an obsolete “hypothetical” language (a parody of the English private school system). The people of Erewhon are all uncommonly beautiful, due to the fact (I assume) that a process of eugenic selection has taken place over generations – as the sick are destined to perish in prison. The society of Erewhon maintains a surprising equilibrium, despite the violation of the central conventional values of Higgs' society.

Book of Machines, in Erewhon (illustration Robert Gibbings?)

The citizens of Erewhon, fearing the rapid advance of innovation – and, it seems, following the same reasoning as the author of Darwin Among the Machines – had stopped all technological development some centuries before our hero’s arrival. In a revolt against technology the people “wrecked all the more complicated machines, and burned all treatises on mechanics, and all engineers’ workshops”. The Erewhonians retained a museum of wrecked machines, which serves as a warning against the hubris of technological progress. They also left a text, The Book of the Machines, which gives an account of the reasoning behind the ban. In Erewhon the text is recounted by Higgs. The Book of the Machines presents a treatise on consciousness – built and begotten – which binds together the discourse of the machine and discourse of evolution.

Butler comes straight to the extreme implications introduced by Wallis’ in 1858: “But who can say that the vapour engine is not a type of consciousness?” asks the author of The Book of the Machines “Where does consciousness begin and where end? Who can draw the line? Is not everything interwoven with everything?”

The author goes on to challenge a series of divisions between organism, environment, machine, and consciousness which are problematised by the arrival of the “self-regulating” machine controlled by negative feedback – the vapour engine. The adaptations performed by the steam engine are not designed, in the sense that a consciousness outside of itself dictates its adaptation. Any alterations and adaptations are dictated within itself.


In Evolution Old and New (1879) Butler’s argument extends the dialectical turn established in Erewhon. Butler points out that the modern discourse of the machine is produced within the discourse of evolutionary theory. This is because central to the discourse of evolution is the issue of design. The issue of design (purposeful or otherwise) was not adequately accounted for in the dominant Darwinian evolutionary theory of Butler’s time – where the agent of change is simply chance.

For Butler, when one addresses the idea of evolution the issue of purposeful design inevitably becomes central. Whether you take the position that nature is designed or that it is ruled by chance, the metaphor of the machine is inevitably played against that of the organism. Even theists who would argue against natural selection would characterise the supreme deity as a supreme designer (as had been the position of William Paley’s Natural Theology (1809). Butler’s Evolution Old and New examines this dialectic in some detail.

To briefly recap, for Butler, in Evolution Old and New the discourse of purpose is entwined with the discourse of evolution because the discourse of evolution inevitably invites an analogy between the human designer and the divine architect. One side of the argument allows that the complexity evident in nature is proof of a higher purposeful entity. The counter to this position is that change and adaptation are the product of chance. Butler opposes both these positions. For Butler, the evolutionary argument produces a distinction between the “mechanical” and the “vitalistic”. Butler goes to great lengths to point out that this distinction is fundamentally false. Butler argues against an interpretation of Charles Darwin’s natural selection which over-emphasises the role of chance in adaptation, in favour of a model of adaptation closer to Charles’ father Erasmus Darwin, and the elder Darwin’s near-contemporary Jean-Baptiste Lamarck.

Butler argues that adaptation is a form of non-conscious learning, an embodiment of a series of differences within the organism over generations. In challenging the emphasis of chance in evolution, and in affirming a high degree of complexity and difference, Butler attempts to articulate a phenomenon yet to be named: negative entropy.

In Evolution Old and New Butler poses the question: if adaptation is a matter of chance, given the order and complexity within nature, would it not be fair to suspect that the dice has been loaded in some way? To refute such a position, which would invite a form of “vitalism”, Butler introduces the notion of teleology (purpose) and invokes the theories of German evolutionist Ernst Haeckel (if only in order to dismantle them at a later point in the argument).

Butler, outlining Haeckle’s premise, identifies two distinct approaches “a causal or mechanical” approach and a “teleological or vitalistic” approach. Up until the advent of evolutionary theory the purposeful and vitalistic was the auspices of the biological world. Nature and all the organisms within it were driven by purpose and ruled by design. It is plain therefore that there must either be a designer who "becomes an organism, designs a plan, &c.," […] “or that there can be no designer at all and hence no design”.[7] Butler takes exception to this latter position that “purpose in nature” does not exist (which essentially was Haeckle’s), but he also rejects the necessity for a divine hand – which must itself be the product of purpose and design and must resemble the organisms which it designs – which sets up an eternal regression which will never reach first cause.

So, I need to clarify what Butler means by purpose.

Purpose, for Butler, relates to adaptation – a thing fulfils a purpose when it is adapted to a particular end. But there are different types of adaptation. Butler takes as his examples a corkscrew, which is man-made and adapted to a particular purpose. He puts this against a pair of human lungs, which have adapted through natural selection. For Butler “the very essence of the ‘Natural Selection’ theory, is that the variations shall have been mainly accidental and without design of any sort, but that the adaptations of structure to need shall have come about by the accumulation, through natural selection, of any variation that happened to be favourable.”

Here Butler establishes a definition of purpose which accords with the definition developed in Wiener et al’s Behaviour, Purpose and Teleology (1943) (see chapter one), and it is built on exactly the same technical standard: that a new definition of purpose is established through the operations of a servo-machine, in which negative feedback is the regulator. It is with this technical standard that Butler (and later Bateson) will apply to the regulative principles of evolutionary adaptation.

In Butler, a structure is established through an accumulation which favours ordered variation. With this Butler sets out the aim of Evolution Old and New, to reintroduce the notion of purpose into evolutionary processes without necessitating a deity as its grand designer or leaving the issue to arbitrary chance. To establish a ground for such a discussion, Butler introduces a hypothesis very close to negative entropy: that it is the accumulation and reiteration of information within an organism which favours variation.

To establish his opposition to conscious design, Butler quotes extensively from William Paley’s Natural Theology, or Evidences of the Existence and Attributes of the Deity (1802), which, in order to prove conscious design within nature, describes the anatomy of the human body in mechanical terms.[8] The bones and joints are like delicately engineered hinges and pivots; the eye is far more sophisticated than any telescope; the most rudimentary human organs far more complex than any clock. Paley goes on to describe the exquisite complexity of the human circulatory and nervous system, in all its forensic detail, as proof of conscious purpose. Paley, in his day, was attempting to refute the theories of Erasmus Darwin by arguing – as people would of his son Charles’ work – that the sheer complexity and order of the natural organism is evidence of a purposeful designer.

Butler takes up this idea and develops the argument to a significant degree, returning repeatedly to the machine juxtaposed with the organism, which had been a central figure in Butler’s argumentation since the days of Darwin Among the Machines. One could, for instance, present a human foot in its complexity and detail alongside a man-made foot composed of ingenious mechanical parts:

“We not only feel that there is a wider difference between the ability, time, and care which have been lavished on the real foot and upon the model, than there is between the skill and the time taken to produce Westminster Abbey, and that bestowed upon a gingerbread cake stuck with sugar plums so as to represent it, but also that these two objects must have been manufactured on different principles.”[9]

Butler describes these different principles by pointing out that a purposeful designer, when designing a machine or building for a specific function, would not go through redundant generations of design. The designer would not first need to erect a hut before designing a cathedral, for instance, and yet nature seems to do precisely that. Closer to the cybernetic era, Kenneth Craik will describe this ability of the human mind to model as a way in which human beings are able to conserve energy through the storage and transmission of information, Craik’s The Nature of Explanation (1943) describes such negentropic reactions as “down-hill reactions” which “saves time, expense and even life”[10] (I will come back to Craik’s ideas later).

In contrast to the purposeful, down-hill, design of human beings, Butler continues “One would say that nature feels her way, and only reaches the goal after many times missing the path.” To ease Butler’s argument into the cybernetic era I will describe the “path” as a threshold or perimeter and the “many times” as the many iterations which eventually find their “goal”. The structure of such a “path” and the presence of the “goal” are both immanent within the system.

Back in Evolution Old and New, Butler is in a heated imaginary debate with the theist, Paley. Butler demands that Paley produce his designer of nature and asks what form such an entity could take. In Butler’s imagination Paley retorts with an equally vehement demand of habeas corpus: “where is your designer of man?” Butler’s answer draws on the distinctly cybernetic interpretation of purpose we outline above. For Butler it is man himself who is the artificer, “best fitted for the task by his antecedents”. This is not a particular man or a man from any given generation but rather man (humanity) over the entirety of its existence. Butler describes the organism of the human adapting in co-extensive relation with it's environment as follows:

“[…] we say that the designer of all organisms is so incorporate with the organisms themselves so lives, moves, and has its being in those organisms, and is so one with them they in it, and it in them that it is more consistent with reason and the common use of words to see the designer of each living form in the living form itself, than to look for its designer in some other place or person.”[11]

Here Butler sketches a picture of the process of biological adaptation which will be familiar to readers of Bateson. In Butler and Bateson’s vision of adaptation the distinction between environment and organism (man + environment) is arbitrary, there is no precise point at which the organism (human) is wholly distinct from the environment. This is also the vision handed down to Butler by Dr. Erasmus Darwin, Comte de Buffon (Georges-Louis Leclerc), and Jean-Baptiste Lamarck, which does not deny purpose, or demand a designing hand outside of the organism itself:

“It is that the design which has designed organisms, has resided within, and been embodied in, the world of organisms themselves.” […] “being that power in [organisms] whereby they have learnt to fashion themselves, each one according to its ideas of its own convenience, and to make itself not only a microcosm, or little world, but a little unwritten history of the universe from its own point of view into the bargain?”[12]

Butler lacks the word "genetics" – which will be introduced into the English vocabulary by William Bateson, (Gregory’s father, see Vibrations). Whilst Butler’s proposition resolves the distinction between purposeful design and chance, it raises a question which was at the heart of the evolutionary debate: what agent, or “power”, facilitates such changes? How could Butler articulate the function of homeostasis within a discourse in which the gene (a unit of heredity and also a unit of probability) and feedback are absent?


Referring to an earlier work, Life and Habit (1877), Butler describes a process of evolution in which ingrained activities are reinforced on a local level, even after such a time that they are redundant. For example, we are oblivious to the function of our internal organs and give them no more thought than we would to

“Julius Caesar in the month of July” […] “we have made the one kind [of organ] so often that we can no longer follow the processes whereby we make them, while the others [exterior organs] are new things which we must make introspectively or not at all, and which are not yet so incorporate with our vitality as that we should think they grow instead of being manufactured.”[13]

Again, Butler draws our attention to the false dichotomy between the mechanical and the organic and at the same time to the equally false dichotomy between the interior and the exterior, between organism and environment. Such distinctions are reinforced in order to maintain order on a local level. This is resonant of Gregory Bateson’s application of the notion of levels of abstraction, and the debt is acknowledged in Bateson’s Steps to an Ecology of Mind. On the matter of levels of the mind, notes Bateson: “Samuel Butler's insistence that the better an organism ‘knows’ something, the less conscious it becomes of its knowledge, i.e. there is a process whereby knowledge (or "habit" — whether of action, perception, or thought) sinks to deeper and deeper levels of the mind.”[14]

Habit, in Butler, might be understood as a form of unconscious code or embedded knowledge which serves to conserve energy and transmit information within the organism. Habit traverses both unconscious and conscious action. Gregory Bateson states:

“Samuel Butler was perhaps first to point out that that which we know best is that of which we are least conscious, i.e., that the process of habit formation is a sinking of knowledge down to less conscious and more archaic levels. The unconscious contains not only the painful matters which consciousness prefers to not inspect, but also many matters which are so familiar that we do not need to inspect them. Habit, therefore, is a major economy of conscious thought.”[15]

Bateson recognises habit as an expression of negentropy, but for Butler, as with Bateson, these habits might prove to be counterproductive. Butler’s evolutionary theory and theory of habit can be understood as antecedent to Gregory Bateson’s ecological theories. The notion reappears in Bateson’s essays from the late 1960s and early 1970s, in which the different levels of (un)consciousness are understood as variables which allow flexibility or re-enforce inflexibility within a system (as we will see later). [16]

It is on the level of genetic theory, which Butler was in great difficulty of articulating adequately in his day, that Bateson outlines Butler’s position, which is that nature relied on cunning rather than luck (chance), and explains how more recent knowledge of genetics had resolved the deficiency in Butler’s theory.[17] Bateson is also able to resolve Butler’s unfinished business through an understanding of the role of negentropy in evolutionary development, at the heart of which is the organisation of information.

Bateson: “The notion is that random changes occur, in the brain or else-where, and that the results of such random change are selected for survival by processes of reinforcement and extinction. In basic theory, creative thought has come to resemble the evolutionary process in its fundamentally stochastic nature.”[18]

In which case, Bateson maintains a little later “Samuel Butler's hunch that something like ‘habit’ might be crucial in evolution was perhaps not too wide of the mark.” [19]


In the pages of Erewhon Butler makes his most visionary claim in the starkest terms, looking toward Grey Walter’s cybernetic tortoise, toward Norbert Wiener’s Cybernetics and toward Bateson's ecology of mind. The narrator of the Book of the Machines writes:

“Let anyone examine the wonderful self-regulating and self-adjusting contrivances which are now incorporated with the vapour-engine, let him watch the way in which it supplies itself with oil; in which it indicates its wants to those who tend it; in which, by the governor, it regulates its application of its own strength;[…]” but “man” must "be awakened to a sense of his situation, and of the doom which he is preparing for himself.” [20]

Butler’s relation to the machine and organism is similar to the double-sided relation between organism and machine outlined in Wiener’s Cybernetics (1948) and The Human Use of Human Beings (1950). What matters for Butler is not the distinction between the machine or organism per se but rather the machine or organism’s relation to purpose. Long before Wiener Butler identified the coming discourse as the discourse of the machine. Long before Jacques Lacan would outline the dialectic of the machine (in his 1954-55 Seminar II) – Butler would describe his own dialectic between the machine and organism, a dialectic which was produced within the discourse of evolution. The knowledge of a progressive ‘design’ or apparent purpose within nature first produces an uneasy equivalence between the designs of man and the designs of nature – the contradictions of both are exposed, brought into the open. This dichotomy is transcended by a new understanding of purpose, and Butler proposes an unconscious purpose which is itself embodied in the act of adaptation within evolution. Here Butler outlines a dialectical relation which Wiener, Bateson, and Lacan would echo in the twentieth century.

  1. S. Butler, Notebooks (1885)
  2. G. Bateson, STEM 434
  3. Specifically; S. Butler: Darwin Among the Machines (1863); Evolution Old and New, Or the Theories of Buffon, Dr. Erasmus Darwin and Lamarck, as Compared With That of Charles Darwin (1879); Erewhon (1872) and Butler's Notebooks (1885)
  4. S. Butler, Darwin Among the Machines, The Press, 13 June 1863, Christchurch, New Zealand,
  5. S. Butler, Darwin Among the Machines, The Press, 13 June 1863, Christchurch, New Zealand.
  6. Alfred Russell Wallis: On the Tendency of Varieties to Depart Indefinitely From the Original Type (S43: 1858)
  7. S. Butler EO&N
  8. William Paley, Natural Theology, or Evidences of the Existence and Attributes of the Deity (1802)
  9. S. Butler EO&N 5
  10. Edward Craik The Nature of Explanation (1943) Cambridge
  11. Butler EO&N 30
  12. S.Butler EOaN 31-32
  13. S. Butler EOaN 157
  14. StEM 151
  15. StEM 151
  16. StEM287
  17. Bateson StEM 260
  18. StEM 260
  19. StEM 263
  20. Samuel Butler, Erewhon p.220