Introduction: 2003


What exactly was behind all that stuff about ‘observation languages’, ‘elementary propositions’, ‘basic statements’, and ‘protocol sentences’ once found in Wittgenstein and Carnap et al? The idea seemed to be that if you only simplified things enough and got back to the nitty gritty of bald description (or somehow got back to the ding an sich through bald description) then language wouldn’t lead you astray. But as any cameraman might have explained, if you’re really interested in descriptive accuracy, then for very large reaches of the known universe, both micro and macrocosmic, a photon map beats verbalising any day.


As for information about the unknown universe—photography is much better than words. It’s a pity that Karl Popper, who clarified a number of things, carried his hostility to the “bucket theory of the mind” and to naïve empiricism so far he would probably not have understood that bottomless information buckets can be very useful. He believed that selective perception sabotaged human observation at every point. In a typical statement he says that “most dissectors of the heart before Harvey observed the wrong things—those which they expected to see. There can never be anything like a completely safe observation free from the dangers of misinterpretation.” [Conjectures and Refutations, page 41, footnote 8.]


No doubt this is so. But real-time photon maps can show you the circulation of the blood by staining techniques (whatever you may believe and however determined you are not to see it), just as they can record the activities of particles in cloud-chambers (which no-one can either see or predict) and show the vortices of sunspots or the explosions of distant super-novae. Theory of course must design and cast the net; but what is caught is something else again. In informational terms, man proposes: photography disposes. For countless areas of countless branches of the natural sciences (not the social sciences, where consciousness and feedback supervene) mechanical description replaced linguistic description long ago.


Such, anyway, were the ideas behind this free-wheeling rumination on aspects of Wittgenstein’s Tractatus along with Popper’s theory of ‘World 3’ as set out in Objective Knowledge. The essay first appeared in Art International in January 1978. Since that time both the reach and grasp of what I call ‘objective graphics’ (which supersede human sensoria for observational and record-keeping purposes) have continually expanded until now there is hardly any aspect of life free of electronic eyes and ears. Furthermore, they are valued precisely because they work like buckets, catching and holding far more evidence than is sought twenty-four hours a day.






Epistemology Without a Knowing Cameraman


by Roger Sandall


[Originally published in Art International, January 1978]



"It frequently happens, moreover . . . and this is one of the charms of photography . . . that the operator himself discovers on examination, perhaps long afterwards, that he had depicted many things he had no notion of at the time." Fox Talbot, pioneer photographer, 1844.1

Among the defining criteria of the objective knowledge of World 3 is the "possibility or potentiality of being understood, (the) dispositional character of being understood or interpreted . . . And this potentiality or disposition may exist without ever being actualized or realized". Karl Popper, 1973.2

"2.131: In a picture the elements of the picture are the representatives of objects." Ludwig Wittgenstein, 1922.3





Fox Talbot's modest view of the photographer's role belongs plainly to another time and place. How things have changed. The nineteenth-century's humble "operator" working quietly away in the decent obscurity of his darkroom has long since given way to the strident practices of "creative artistry", and it is not to be expected that an ambitious creative artist would acknowledge, much less admire, those accidental impingements of an objective world which the genius of the photographer "had no notion of at the time". In the most famous examples of photographic creativity nature herself plays a secondary role. And as for the dumbly mechanical nature of photography, the less said about that the better.The main reason for this state of affairs is embarrassingly obvious: it is the historic deference of "rude mechanicals" for "art". Few artisans more than photographers have felt so keenly and so long the pressure of an older and more prestigious tradition. There is scarcely a primer on lensmanship which doesn't pointedly emphasize, sometimes on the very first page, that taking a picture "is not just a mechanical act". There is scarcely an introduction to an album of photographs which doesn't go out of its way to celebrate the role of the interpretive human eye behind the uninterpretive, glassy lens.


In recent years this endemic subjectivism has been given a huge boost by the cult of self-expression. No opportunity has been lost to equate the taking of pictures with the making of pictures (the second, of course, being the more admirable), as if the artist's ego felt directly imperilled by any recognition of the essentially objective, physico-mechanical nature of photography itself. Underlying this, needless to say, is the assumption that mental states are self-evidently good, whereas machines are self-evidently bad. Ergo, if I use a machine like a camera I must justify this use in terms of its thorough subordination to mind; I must show that what science imagines to be an instrument of improved perception is more wisely seen as a self-expressive device.


All of which this particular cameraman finds somewhat perverse. The struggle to distinguish what is inside our heads from what is not has been embarrassingly long and difficult; and in this struggle, photographs (along with tapes as well; together they make up the family of "objective graphics") have played a very useful role. Just think, for example, what progress might have been made if cameras had been around in the 18th century. Would Sam Johnson, refuting Berkeley's subjectivism, have kicked a stone or taken a photograph? Given his natural boisterousness he might easily have done both. At all events there's not much doubt that he would very soon have noticed what Fox Talbot noticed over a century ago—that the world of objective graphics is full of surprises. Not only does the operator never get quite what he expected, in most cases he gets more.[i] That of course is precisely why film and tape are now used for both scientific enquiries and social investigation, from radio-telescopic explorations of space and the mapping of x-ray nebulae, all the way down to the police tape recordings of witnessesstatements or last night's television news.


And this has profound implications. Today the more vulgar forms of irrationalist epistemology shelter behind two protective doctrines—the Doctrine of Selective Perception and the Doctrine of Ideology. The first declares us to be incurably one-eyed; the second accuses us of only using that eye which serves our own interests, narrowly defined. But the interesting thing about objective records on film and tape is the way they relentlessly undermine these doctrines by their very nature. An incorrigible ideologue soon finds what he's up against. The camera's sheer inclusiveness works against him, and the only way he can make its images say what he wants them to say is by the most determined, and visibly eccentric, editing. Negative evidence for this can be found in the total absence of modern documentary film-making in those vast political regions where they take ideology really seriously—between the River Elbe and the shores of the China Sea. For it is not always understood, and perhaps needs stressing, that the technology of social enquiry does not exist in a vacuum: It presupposes a social philosophy which allows and encourages enquiry to take place.


Is it possible to pursue this question further? Can we look beyond the social arrangements which allow the technology of empirical enquiry to exist, and examine some of those ideas which, if they have not actually furthered, have at least accompanied its rise? After Marshall McLuhan's extravagances it has become harder than it should be to argue the less apocalyptic connections between technology and oth­er areas of life and thought. But this doesn't mean that they don't exist; and to the objectivist more interested in the kind of thinking which relates thought to things rather than to sensations (let alone the stoned sensoria of mass society), there may still be a few lessons to be learnt.





In order to draw as clear a distinction as possible between objective and subjective graphics Fox Talbot's observation is a good place to begin, especially that scene of a photographer dusting off an old print and being delighted to find all sorts of things in it he didn't know were there. This experience, while common enough in photography, is even more common in documentary films, where individual frames may be examined arid found to contain far more than the cameraman was conscious of at the time. By way of contrast, you can't imagine Bruegel the Elder dusting off an old canvas and finding a dog or a steeple he was unaware of painting years before. In subjective graphics the painter inspecting his past work finds only what he expects to find—or if he doesn't, he attributes the unwanted trouvaille to a failing memory of events. This predictability is what makes subjective graphics "closed systems" in a solipsistic sense, and while on the one hand it helps to make them more reassuring (which is one legitimate function of art), it also guarantees a certain epistemological sterility on the other. An exhaustive list of everything stored in the memory of the painter's mental cupboard will tell you all that the work of art has to say: the state of affairs it represents is a state of mind.


Now just the opposite is true of objective graphics. You could have the fullest possible inventory of everything in a photographer's mental cupboard and it wouldn't help you a bit. This is obvious enough in the case of the sort of electronic scanning equipment used in radio-telescopy which is, by definition, searching the sky for things which may or may not exist. But although less obvious, it is also true of the cameraman shooting film for the evening news. Cameramen are only human, and it must be confessed that not all of them are single-mindedly absorbed in their work. It often happens that their attention swings to and fro between whatever can be seen in the viewfinder and such normal human preoccupations as food, and sex, and vengeance. It might even be argued that a highly skilled cameraman should be able to give his mind almost entirely to the latter; for it is one of the marks of his skill that after long experience he knows so exactly what the producer back at the studio regards as "news" that he is able to frame it and film it almost subconsciously.


Of course news film is comparatively demanding: certainly the work keeps you awake. But when you turn to something like a cricket match on a drowsy summer's day then the cameraman's mind is bound to wander; while sound recordists, especially during\ interviews, are always falling asleep on the job. Yet despite the mental vagaries of their operators, cameras and sound recorders (no less than radio-telescopes) continue faithfully imprinting their maps of states of affairs. And like Fox Talbot's photograph, the resulting mechanical maps possess "the dispositional character of being understood or interpreted" as Popper puts it, whether or not this ever actually happens, and whether or not there is any available consciousness to do it for any number of years.5


To speak of objective graphics as maps, and maps, moreover, containing coordinate points, is plainly to use the word map in an extended sense. The extension however is not mine alone: it has already been made by those astronomers who in recent years have provided us with pictures of radio-galaxies and supernova remnants—pictures which they call maps.6 And the interesting thing about this usage (and about the images themselves) is that it brings out the generally hidden nature of photographs and tape recordings as patterns of coordinate points too. In an x-ray map of a nova the sky has been scanned to detect varying intensities of x-ray emission; and the finest discriminable units of emission are then displayed as a chequered pattern of variously coloured squares. It is true that the colours are arbitrary and reveal more about the responsible technician's aesthetic taste than anything else. But as in all objective graphics the position of the squares is emphatically not arbitrary. In these astronomical maps each little square (or "ele­ment of the picture" in Wittgenstein's phrase) is a coordinate point defined as so many seconds of arc in a given sector of sky; and furthermore, there is a close resemblance between these squares and what Witt-genstein, talking about linguistic maps, referred to as "the representatives of objects".


Now although it is a long way from the grand cosmic imagery of a super-nova to the family snapshot taken last summer, it is their shared character as maps, maps made up of definite coordinate points formed by electro-magnetic waves, which certifies their objectivity. In the case of ordinary photographs of daily life we usually attend so exclusively to form and content that the underlying principle of coordinate pointillism is overlooked. Yet that is how they are made. A light ray emanating from a spot in the world follows a mathematically fixed path through a lens and ionizes an atom of silver in the photographic emulsion. Simultaneously a neighboring ray ionizes a neighboring atom, making two. And so on ad infinitum until all the particles in the emulsion have been either struck or not struck by rays of light. In the final event what we get, multiplied a million-fold, is a map consisting of a myriad minute particulars, each ionized atom (or developed crystal) a point in a coordinate graphic system corresponding to the world beyond. In recordings on magnetic tape, whether sound or video, we find the magnetization of iron particles instead of ionized atoms of silver: but in each case what the patterns of particles amount to is a system of coordinate points. Both ionized atoms and magnetized particles are "representatives of objects" which map the physical world.





In discussions of this kind the word "representation" seems impossible to avoid. This is unfortunate because it is all too easy to confuse the pictorial and the political meanings of the term; indeed this confusion is almost inevitable since their meanings are very nearly the same. Political representation proposes that a man be said to represent a district, a class, or a nation, when he synecdochically "stands for" numerous other human elements in the political map. It embodies, in brief, an organizational economy of time and motion. According to the pictoral meaning, on the other hand, a few strokes of paint may "stand for" trees or towns or people. This embodies an economy of meaning and display.


And for a perfect example of the way these meanings blurr together into one we need look no further than the quotation from Wittgenstein already cited: "The elements of the picture are the representatives of objects." Reading this, one all too easily conjures up an image of a kind of pictorial House of Representatives full of splashes of paint and spots of colour in various shades of red and blue. With this in mind it becomes as much an ethical as an artistic necessity to know upon what principle a splash of paint represents a collection of men, and by what right the painter decides that it shall. There is a decided risk here both of burying the argument under a metaphor, and of politicising the whole discussion; nevertheless it may as well be stated that in objective graphics the process of "representation" has much more of a particularistic, grass-roots character built into it than it does in either painting or politics, for its history has been guided by the ideal of one man/one vote/one representative. Under the normative goal of fidelity, and of higher and still higher fidelity, objective graphics continually strive toward an ideal in which each discriminable unit of space/time in the universe should be represented by a discriminable unit in the photograph or the recording tape as well.


Fidelity and faithfulness are qualities which are less admired today than previously, but in the case of objective graphics a forthright and unembarrassed use of both terms should be encouraged at all times. This is partly desirable for historical reasons, since as far back as 1828 we find Nicephore Niepce, the man who first obtained and fixed a camera image, proclaiming his intention to "copy nature with the greatest fideli­ty".7 But it is also appropriate because, despite an uncompromisingly technical foundation, and a vocabulary of things like "linearity" and "time-base stability" and "frequency response", the precise modern concept of high fidelity really incorporates the older, vaguer, nineteenth century one. The effect of measures like frequency response is to make exact and explicit the implicitly quantitative nature of old-fashioned "exclamatory" fidelity—what might be called the fidelity of the "aha!" response. The "aha!" response is determined by the frequency response, and it can be heard today every time two photographs or films or radio-telescopic pictures or sound or videotapes are compared and one of them is found to contain incomparably more and clearer detail than the other. In objective graphics the Law of More is the law of their evolution.


The Law of More insists that if an astronomer is given a choice between two radio telescopes he is bound to prefer the one giving most detail, just as it requires that any ordinary photographer should choose the sharper of two lenses. Whether the available detail is used is another matter. It is of course true that the astronomer may elect to display his data in deliberately coarsened aggregates, just as the local photographer may not always want to use the finest resolution of his lens in the final prints. To take a famous example, when Julia Margaret Cameron softened the focus on such portaits as her head of Sir John Herschel, turning the astronomer's hair into a corona of solar flares, she pioneered a technique used by numerous cameramen to spiritualise the flesh of Hollywood stars.8 But this only goes to show the mutual antipathy of objective and subjective graphics. And in the resulting fog of soft-focussed spirituality it is important not to lose sight of the main issue. This is that the capacity of one instrument to get more detail than another, and the "disposition" or "potentiality" of these details to represent objects (whether or not this capacity is realised), is what guarantees that it will be the most sought-after instrument in its field. For a man who wants to find out what there is in the world, rather than what he has in his mind, the most highly prized instrument is the one which repre­sents with the highest fidelity physical states of affairs.


A physical state of affairs always exhibits varying degrees of order, more or less; and physicists sometimes say that degrees of order and quantities of information add up to much the same thing.9 This is directly related to the Law of More which governs the evolution of objective graphics, for what the constant­ly multiplying coordinate points in higher and higher fidelity images are actually coordinated with (or what they correspond to), are the degrees of order percep­tible in the world. When information is synonymous with order it is measured in "bits", those anonymous minimal units which enable us to distinguish, to discriminate, to tell features from featurelessness, and something from nothing (or some thing from no thing). As narrowly interpreted by a biologist the Law of More might mean more detailed information about organelles; as interpreted by an astronomer it might only mean more detailed information about galactic boundaries; but in its broad and general form applied to all objective graphics indiscriminately the Law of More always means more "bits". From which it follows that what the "ahas" of the exclamatory criterion of fidelity really celebrate is the ability of objec­tive graphics to distinguish finer and finer degrees of organization in the physical world; to map it, as it were, with closer and closer coordinate perfection.


These considerations should further sharpen our distinction between objective and subjective graphics, for in the realm of art the degree of order reflects the order in the artist's mind. There, as Gombrich was at pains to show, norm governs form;10 and the vocabulary which elaborates on this relationship is one which richly exploits the possibilities of Symbols and Meaning and Sense. No one who has enjoyed using such terms could be insensitive to their loss, or unaware how poor a substitute for such heady notions a diet of anonymous "bits" of "information" must seem. To insist that a professional appreciator of subjective graphics should forgo this prestigious and delightful terminology in favour of mere "bits" would be to dash from his lips a beaker full of the warm south and replace it with a mug of gruel. Meaning, after all, is sought by poets; whereas information is more often sought by the police. But there it is; and as a true prophet of the Reformation of Objective Graphics ich kann nicht anders. Besides, a renunciation of Symbols and Meaning and Sense will make it a lot easier to relocate Wittgenstein's picture theory of meaning where it naturally belongs—in the world of objective graphics.





It would not be too hard to find people who would agree that of the two Wittgensteins, Wl and W2, it is the objectivistic thought of the Tractatus which is more scientifically significant. But even among these there would be few defenders of the "picture theory of meaning". Wittgenstein's early belief that language pictorially maps reality, and his attempt to define the conditions of exact linguistic cartography, are generally regarded as woefully misconceived. No doubt they were: but then it is precisely because language can never map reality with anything approaching coordinate precision that objective graphics are important—they do it so much better. And when one ponders on those parts of the Tractatus one can understand it is hard to avoid being struck by the ways in which Wittgenstein's criteria for linguistic representation, and his atomistic criteria for fidelity, fit much more naturally the representation of objective graphics. "A picture is attached to reality" we read at 2.1511.11 "It reaches right out to it" with what seem to be feelers connecting words and things. Like most people who have never actually seen a Wittgensteinian picture I rather doubt if this is so. But how much more sense the same idea makes in the case of objective graphics, for exact ontological maps inhere in the very process by which they are made, projected and drawn by electromagnetic waves which do indeed "reach right out" from the reality they project.

Wittgensteinian fidelity even has a numerical criterion which relates to the Law of More. At 4.04 we read that "in a proposition there must by exactly as many distinguishable parts as in the situation that it represents".12 Now the Law of More requires that for any given field of view or state of affairs, cosmic or microcosmic, the more faithful report is the one which provided more discriminable bits. And by making only a few substitutions to 4.04 we can see how easily it can be adapted to our purposes: "In an objective graphic there must be exactly as many bits as in the situation that it represents." If there are not exactly as many bits then the fidelity will be lower, and the information inherent in the state of affairs will not be faithfully represented. Lest some of these speculative interpretations be thought too whimsical to be taken seriously I should perhaps add that there is more to 4.04 than the short excerpt given above. In a second paragraph Wittgenstein reinforces his notion of numerical equivalence by adding that a proposition and a state of affairs "must possess the same logical (mathematical)  multiplicity.  (Compare Hertz's Mechanics on dynamical models)". The reference to the famous physicist is significant: he was a man who had thought much on the nature of faithful representation. And it is appropriate that the quantification of fidelity has become inseparable from his name; wherever you find audio systems in handsome cabinets of teak veneer men talk about hertz or kilohertz to establish the fidelity with which their systems represent.


Yet it is perhaps an earlier thought in the Tractatus which most clearly relates to that essential fea­ture of objective graphic representation—its coordinate nature. At 2.15 we find that "the fact that the elements of a proposition are related to one another in a determinate way represents that things are related to one another in the same way".13 Points which relate to each other in a "determinate way", corresponding point for point with reality, are really coordinates. And once this is grasped it becomes apparent that Wittgenstein's "picture theory of meaning" is equally, and perhaps more usefully, a "coordinate theory of representation", a rechristening which preserves his general aim of establishing certain principles of analysis while avoiding all reference to symbols and meaning and sense. "In an objective graphic the elements of the picture are the representatives of objects." And in the sense that they represent them, they are indeed.


Which brings me to that outstanding example of musical low-fidelity, Wittgenstein's gramophone record; or rather, the serial stages of music he describes from idea, to score, to performance, to the objective graphic of the mechanical recording. After all the plain good sense about mapping and coordinate correspondence, what he has to say at 4.014 is painfully disappointing. Despite the importance attached to numerical equivalence and to the correla­tions of parts which exact correspondence requires, he seems casually to imply that there's really not much difference in the representational fidelity of machines and men:

"A gramophone record, the musical idea, the written notes, and the sound-waves, all stand to one another in the same internal relation of depicting that holds between language and the world."14

Now it's one thing to say that they share some general internal relation, and another thing entirely to imply an external equation. Yet this is what soon follows:

"4.0141: There is a general rule by means of which the musician can obtain the symphony from the score, and which makes it possible to derive the symphony from the groove on the gramophone record, and, using the first rule, to derive the score again."


I suppose much depends on how precisely we interpret the word "score". But the gist of what he says is not only at odds with my general argument here; it also contradicts his own numerical and structural criteria for facsimiles. The distinction between what happens when a musician reads a score and when a needle reads a groove is absolute. We have high-fidelity amplifiers and we value them. We don't have high-fidelity violinists and no-one would pay to hear them if we did. What we look for in violinists are those felicitous infidelities of style which would be the despair of anyone who wanted all performances of a score to sound the same. These infidelities ensure that one performance is never a coordinate representation of another.


Introducing a human performance governed by general rules of interpretation into any chain of mechanical replications is the surest possible guarantee of low fidelity. And this particular example of Wittgenstein's early thought is not only significant for the way it dramatises the contrast between objective and subjective graphics; it also forms an intellectual bridge between the hard-edged objectivism of the Tractatus and the notably soft-edged subjectivism of his later years. It's a long way no doubt from the subjective degradation of musical performances in the Tractatus to the preoccupation with rules and games in the Investigations; and from there to the modish wisdom of the academicians of ideology. And perhaps the only excuse for mentioning them all in one breath is to suggest that once you lose your grip on the idea of description as fidelity to states of affairs, a fidelity expressed in quanta of information, then very little stands in the way of the arrant subjectivism of the present day. The ruminations of the Investigations relativise truth into linguistic localism; in the doctrine of ideology the point is power.





A recent study of Wittgenstein15 gives some credit to the Austrian psychologist Karl Buhler for redirecting the ideas of the Tractatus away from the picture theory of meaning. It appears that Buhler's view of the mind was aggressively antipictorial: whatever thought and thinking might consist of, images and imagery were mere mental decoration. Instead he proposed something called "imageless  thought" and regarded vagueness in human communication as in­herent, irremediable, and at times desirable. While it is highly unlikely that the author of Objective Knowledge would agree with all of this, Buhler's view of the mind as engaged in a continuous process of eliminative enquiry does appear to have influenced Popper's thought, and Buhler's evolutionary views on language are often mentioned. From the view of the mind as always searching for answers to questions it follows that there is no such thing as the blank reception of data radiating from objects "out there" which is then imprinted on human senses as light waves imprint themselves on photographic film. The mind, Popper reminds us, is a problem-solving device, not a bucket.16


With this proposition no advocate of objective graphics would disagree. Yet the metaphor of the bucket, so inapt as a psychological model, is entirely appropriate as a photographic model. Buckets and cameras are similar sorts of things. Both have large orifices with a habit of indiscriminately sweeping things up, water in one case and light-waves in the other. Of course when Popper rejects the bucket theory of the mind he is also rejecting, and rightly, the photographic theory of the mind as well. But why allow bad metaphors to prejudice one against good machinery? Especially since there is such a strong resemblance between the impersonality of that knowledge he finds stored for an indifferent posterity in “World 3” and the impersonality of the representations of objective graphics, the information they give us about physical states of affairs which always remains as a disposition or potentiality only awaiting the arrival of some Fox Talbotian trouveur who "discovers, on examination, perhaps long afterwards" things which no-one previously had any notion existed. Popper writes about unconsulted books in libraries which contain knowledge in the form of as yet uncontroverted statements about the nature of things; and stresses that these statements are quite independent of any experiencing consciousness.17 Surely much of the information contained in objective graphics is similar?


It's true that one of the things Popper values most as an epistemologist is verisimilitude, and truthful assertions approximating more and more closely to that state. Whereas objective graphics belong to a nonlinguistic and alogical universe of physical representations which show, which display, which mirror—but which do not assert. Sometimes they show with marvellous detail, but for all their precision and exactitude they lack the unique linguistic possibility of negation. Wittgenstein's dogged attempt to assimilate words to pictures in the Tractatus was really aimed at having the best of both worlds; the exactitude of neutral representation and the possibility of propositional affirmation and denial which only belongs in the normative universe of logic. This can't be done. But what pictures can do (and what objective graphics do every day) is provide the necessary observational evidence for propositional assertion. And surely this evidence is part of the objective knowledge of “World 3”? Surely when Rutherford said to Chadwick that all sorts of subatomic activities had been "going on" and expressed surprise that his colleague hadn't known about them he was implying that this was something which Chadwick did know now; or at any rate knew more reliably as a result of the cloud chamber evidence?


It would hardly seem so from a reading of Objec­tive Knowledge, where remarks on the role of special instruments in building up the picture we have of the physical world are few and grudging. In fact one of the only places where the impersonal mechanical fact-gathering is seen in a favorable light is on the occasion when Popper draws on the arguments of that little-known epistemologist, Winston Churchill:


"Some of my cousins who had the great advantage of University education", wrote Churchill, "used to tease me with arguments to prove that nothing has any existence except what we think of it... I always rested upon the following argument which I devised for myself many years ago . . . here is this great sun standing apparently on no better foundation than our physical senses. But happily there is a method apart altogether from our physical senses, of testing the reality of the sun . . . astronomers . . . predict by (mathematics and) pure reason that a black spot will pass across the sun on a certain day. You . . . look, and your sense of sight immediately tells you that their calculations are vindicated . . . We have taken what is called in military map-making a (cross bearing'. We have got independent testimony to the reality of the sun. When my metaphysical friends tell me that the data on which the astronomers made their calculations were necessarily obtained originally through the evidence of their senses, I say 'No'. They might, in theory at any rate, be obtained by automatic calculating-machines set in motion by the light falling upon them without admixture of the human senses at any stage . . ."18


Popper comments: "The argument is highly original;first published in 1930 it is one of the earliest philo­sophical arguments making use of the possibility of automatic observatories and calculating machines (programmed by Newtonian theory)."19


I know of no other place in Popper’s writings where the true significance of objective graphics is recognised. And the basic reason appears to be the one already alluded to—a suspicion that the secret motive of anyone using mechanical analogies is to promote that epistemological enemy, "the bucket theory of the mind". Yet the details of Churchill's statement are certainly illuminating. We have automatic calculating machines, set in motion by light falling on them, and "without admixture of the human senses at any stage". This description exactly fits a considerable amount of hardware now orbiting the earth or voyaging to distant places, hardware which is receiving and storing information on the degree of order in the immediate environment at a rate no mind could cope with, is processing it in the form of coordinates subject to instant recall, coordinates which as in all mechanical graphics are "the representatives of objects".


Of course automatic machinery can be set in motion in more ways than one—sometimes it is triggered when you enter a room. That is how many of the White House tapes were obtained. And it would be hard to find a better example of the discrepancy between intention and use than these unusual examples of modern recording techniques. Presumably President Nixon had the tape-recorders installed in the basement in the hope that he would be laying the foundation of a memorial Nixon archive to be at least as large, if not as pretentious, as the Johnson mausoleum down in Texas. And for months the recorders turned dutifully on and off, building up an impressive store of information about states of affairs which were also affairs of state.


At the end of it all someone went through the tapes and made a number of unexpected discoveries which had been doggedly concealed up to that time. Subjectively they shouldn't have been there at all. Subjectively, they weren't. As the President said to Haldemann, advocating an implausible degree of low-fidelity playback: "You can say I don't remember; you can say I don't recall."20 And it is also true that some of the coordinates were missing. There were long gaps in the tapes where a beleaguered subjectivity had tried stubbornly to mis-represent. But despite all this there was still enough unerased objectivity to ruin the Administration. For—if I may paraphrase Fox Talbot—it is one of the charms of objective graphics that an operator may discover, days, months, or even perhaps years later, that he has done and said many things which all his faculties tell him could never have taken place at the time.



1. W.H.F. Talbot, The Pencil of Nature, 1844-46 (serial). The quotation appears on page 151 of the revised and enlarged edition of Beaumont Newhall’s The History of Photography, published by Secker and Warburg in 1972.

2. Karl R. Popper, Objective Knowledge, Oxford, 1973, page 116.

3. Ludwig Wittgenstein, Tractatus Logico-Philosophicus, Routledge and Kegan Paul, 1961 (translated by D.F. Pears and B.F.McGuinness), page 15.

4. President Nixon, for example, got much more than he expect­ed. The history of the White House tapes exemplifies the dramatic consequences of the data acquisitivencss of objective graphics.

5. A story from the early days of nuclear research at the Cavendish Laboratory nicely illustrates the contrast between the mental state of the operator and the result of his work. We are told that when Chadwick showed Rutherford the first cloud-chamber photographs of recoiling protons, Rutherford exclaimed: "Do you mean to say that all this has been going on and you didn't know it!” (Understanding Physics Today, by W. H. Watson. Cambridge University Press, 1963, p. 16.)

6. See "X-Rays from Super-Nova Remnants", by Philip A. Charles and J. Leonard Culhane, Scientific American, December 1975.


[In the original printing of the essay in Art International the footnotes were unfortunately cut off at this point. The passage from Churchill occurs in his autobiographical My Early Life.]


back to Roger Sandall