Preliminary critique of the concept of the epistemological border (frontière) by Gaston Bachelard

Preface by the student-translator:

I feel that I must make a few brief notes for my reader. First, my reader will be delighted this preface will not be long.

Second, I should like to say that I have made deliberate choices of translating certain words from the French into the English in order to maintain a fidelity to what I perceive to be Bachelard’s fundamental epistemological grounds. That I am noting the deliberate choices — and noting them soon (patience, please) — perhaps testifies to the absence in English equivalent expressions for what appears to be present and resonating more fully, playfully, and spiritedly in the French. For example, the title of the work contains the seemingly easily approached cognate word, frontière. Frontière may be translated into English as frontier, which I imagine many unfamiliar with Bachelard’s work might select for translation. However, frontière may also be translated into English as border. I decided that border was preferable to frontier because the former suggests a human making that the latter does not. That is, frontier suggests there exists an unclaimed landscape of truth waiting only for science to discover it, whereas border suggests and reinforces the erection, the delineation, and the circumscribing acts around what becomes known as the domain of scientific knowledge. Unfortunately my choice of border is unsatisfactory in that the play against frontière is certainly absent in the English, but work must proceed and choices must be made. The other major word choice concerns the translation of the word esprit. Esprit posed problems for me because it can mean both mind and spirit: the French appears to resonate both meanings simultaneously. What to do? I decided that in some expressions, mind was the better option, for others, spirit. But overall, I leaned toward a preference for spirit because mind tends to suggest a certain singularity that spirit does not. It is awkward in English to suggest that several persons — a collective body of scholars, perhaps — all possess a solitary single mind. It is not as awkward to suggest that several persons share a solitary single spirit, especially with respect to accomplishing like tasks and like works for like reasons. Also, as one of Bachelard’s professional translators has chosen to translate one of his works into English with the title, The New Scientific Spirit, I suppose this is not so faulty a decision by me.

Third, all emphasis is from the original.

Finally, I should like to say I hope some of the boldness, playfulness, and spiritedness of Bachelard resonates through my humble translation: After all, these qualities are what make reading a Bachelardien passage so very memorable and marvelous — yes, even in English. So, without further delay, it is my supreme pleasure to present the following 1934 lecture, “Critique préliminaire du concept de frontière épistémologique,” by Gaston Bachelard.


Preliminary critique of the concept of the epistemological border (frontière)

By Gaston Bachelard

Has the concept of the limit of scientific knowledge a sense absolute? Is it even possible to trace the borders of scientific thought? Are we truly confined within a domain objectively enclosed? Are we subjugated to a reason immutable? Is the mind a sort of organic instrument, invariable as a hand, limited as sight? Is it at least constrained to an evolution regulated in liaison with an organic evolution? There are many questions, multiple and connected, which put at stake every philosophy and which indebted give primordial interest to studying the progress of scientific thought.

If the concept of the limit of scientific knowledge seems clear on first sight, it is because at first glance it leans on elementary realist affirmations. Thus, in order to limit the reach of the natural sciences, we objected to all material impossibilities, indeed spatial impossibilities. One says to the scientist: You never spoil reaching for the stars! You never spoil to be sure that indivised corpuscle very well indivisible! This limitation all material, all geometric, all systematic is at the source of the clarity of the concept of epistemological borders. Naturally one has quite a series of very changing, but also brutal, interdictions. We objected to, for example, the impossibility of the triumph of death, to know the essence of life, the essence of the spirit, the essence of the material. Little by little, from a manner more philosophic, one surrounds thought by an ensemble of positions alleged as essentials. In other words, one rejected in the discursive thought the possibility to know the things in themselves and one attributes to a more intuitive, more direct, but not scientific, thought the privilege of ontological knowledge. The partisans of the metaphysical limitation of scientific thought have also given themselves the right to pose a priori from bounds that are without rapport with the thought that they limit. That is true if the obscure concept of the thing in itself is almost unconsciously utilized in order to specify the impossibilities of particular sciences. Thus, the metaphysician repeats: You can speak only of electricity in itself, light in itself, material in itself, life in itself.

But we do not become duped from the false clarity of this metaphysical position. In fact, in order to prove that scientific knowledge is limited, it is not sufficient to show its incapacity to resolve certain problems, to make certain experiments, to realize certain human dreams. It is necessary to be able to entirely circumscribe the field of knowledge, to delineate a continuous limit impassable, to mark a border that truly touches the domain limited. Without this last precaution, one is already able to say that the question of the border of scientific knowledge has not any interest for science. The scientific spirit would be then very capable of taking easy revenges. It would be able to argue that an insolvable problem is a problem badly posed, that an experiment portrayed as unrealizable is an experiment where one places the impossibility of it in the data. Too often the utterance of a limitation implicates a condemnation to failing because the impossible problem already imposes a method of defective resolution.

Insist on this point and we are going to see that the contestation of impossibility is not at all synonymous with a limitation of thought. For example, when one is not able to resolve the squaring the circle, this does not at all prove an infirmity in human reason. This impossibility purely and simply proves that the problem of the squaring the circle is badly posed, that the data of elementary geometry are not sufficient for this solution, which the word quadrature already implicates a misleading method of solution. One must, therefore, allow in the mathematician the task of uttering once more the intuitively badly posed question; one must give him the right to place in a work a method of transcendence suitable to the problem judicially rectified. In order to play each difficulty in succession, one is able to argue from an analogical manner that the problem of death is, as it were, the problem of the squaring the biological circle and that it is, without doubt, very badly posed when one calls for the solution at the level of the human, for example as the upholding of a personality which we have not even in the course of our life a guarantee that it is truly and permanently one. We demand we conserve that which we do not possess. In order to resolve the unsolvable problem of death, one must, no doubt, have recourse to experimental transcendences, to biological transcendences, to the very senses of the mathematician who completes his explanation of material before a new mathematical object.

But in following its adversary over this terrain, the scientific spirit tends only to show it would only need to play beautifully. In reality the debate is not there. This is not about remote and abrupt interdictions that are a good idea to discuss. Science alone is fit to trace its own borders. But for the scientific spirit, to trace distinctly a border, it is already beyond it. The scientific border is not so much a limit than a zone of particularly active thoughts, a domain of assimilation. On the contrary, the border imposed by the metaphysician appears to the scientist as a kind of border neutral, abandoned, indifferent.

It is however very easy to prove that the scientific thought is by essence a thought on the way to assimilation, a thought that attempts transcendences, that supposes the reality before the known and that knows not that it is a realization of its supposition.

Beginning on the most delicate point that offers the most occasions for discussion. Trying in effect to give examples of experimental transcendences. Precisely surrounding this that we understand by that.

Before all, it is necessary to comprehend that contemporary experimentation is, from all evidence, founded on heterogeneity of sensory experience. It is truly too convenient to erase this heterogeneity in proclaiming that after all, all the data utilized by Physics are sensory data. It seems to us, on the contrary, that one must classify the data, that one be also able to demand the experimental data are from the same kind, if they correspond in whole to natural elements. As soon as one clearly formulates this question, one savors that a truly sensualistic position of science is no longer possible. At the very most, sensory qualities and magnitudes are valued by science as the signs of different objective qualities and magnitudes. Experimentation always comes from a domain of observation firstly; so much that one must say that the experimentation rather attempts to contradict the observation that confirms it. In following the historical development of Physics, it would be easy to accumulate examples of transgression of experimental domains. Such would be the case of thermology that attempts kinetic explanations, of optics that attempts mechanical explanations, of chemistry that attempts electrical explanations. That we accept to import only natural phenomena! We will see that scientific thought studies less in analysis than in synthesis; it imposes on itself stranger schemes in its prominent phenomenological traits; it attempts from within to detect the forms by the deformations. In a word, it presents itself as overcoming heterogeneous and primitive phenomena. What remains from the immediate Physics of electricity of the eighteenth century in the electrical science of our epoch? This immediate Physics had hardly even a role in elementary education. If we only contemplate for the first time the importance given to the biological reactions of electricity in science, we will comprehend the decline of the phenomenological immediate. For a long time, the borders of primitive observation have lost all value, not only for comprehending experience, but also in order to give a good definition of basic phenomena. But if the notion of the epistemological border has truly a positive sense, we will see the vestiges in culture; we will find again the structures of observation in the schemas of experimentation. That is not the way it is: the former borders scarcely appeared except as the mark of the narrow-minded; they designate better error than truth.

Naturally we have marked in this brief statement only the points of departure from experimental transcendence. It is there that our thesis is less clear. If we rekindle the opportunity to know the evolution of the contemporary atomistic, the experimental transcendences we will materialize leveled with the grandest clarity. When we return only two years back to where we began. What one returns to accounts for little hope that one has constituted a science of the atomic nucleus! It would well appear that such an atomic border was able to be overtaken, from the lesser nuclear border becoming to be impassable. One feels, for that matter, that these expressions are simply calculated on a spatial schema. Every problem will change on its face if one will capture it from new intuitions, and educate oneself by example compared with intuitions from wave mechanics. We will see, then, that the problems of the atomistic themselves pose not necessarily in patterned form from prohibited regions, from interlocked domains ones in the others, contained necessarily in a center at the transcendental mystery.

We are able, therefore, to assume to demonstrate that experiment transcends observation. Or whenever one transcends the borders of immediate observation, one discloses the metaphysical depth of the objective world. The veil of the Maya is lifted. The intuition critiqued reveals itself as illusion. Here, at once, a confirmation of our rationalistic optimism: The world hidden beneath phenomena is clearer than the visible world. The first noumenal constitutions are more solid than the phenomenal conglomerations. Moreover, the borders of experimentation are in some manner less opaque, less oppressive, than the natural borders of first observation. In effect, if an experiment rectifies an immediate observation, it does it in leaning on the experiment’s coordinates that illuminate one another. To give only one example, as elementary as possible: the fall of bodies, such as it immediately presents itself, is troubled by the resistance of the air to such a point that it itself is disfigured. One will initially make an experiment in standardizing or in eliminating the disturbances. But it illuminates itself when it will take a mathematical form, when we are able to forecast and to extend the mathematical consequences. From a general manner, all scientific progress presents itself as a reduplication of the proofs, as indirect confirmations. The most striking of the verifications are the most indirect. So the coherence finishes by awarding a prize to the evidence.

But coherence is to know to carry out a deepening of the objective experiment, to a point where one is able to speak that there are more possibilities in rational organization than in natural organization. So there are more chemical substances in the laboratory than in nature. Certain chemical bodies created by man are not anymore from a reality than The Aeneid or The Divine Comedy. By certain aspects, it seems to us not anymore useful to speak of the borders of Chemistry than of the borders of Poetry.

It would not be necessary to believe that this indefinite mathematization of the scientific noumenon will be the result of simple abstraction. Recent attempts have been made in order to define scientific phenomena as simple series of effectively realized measurements. Such a method comes back again to confine the scientific thought in a sort of metric phenomenology; to confine science in the reign of the quantity as the sensualistic philosophy will claim to confine the objective thought in the reign of sensory qualities. But those theses of the metrical narrow forget precisely the mathematical transcendent thought that organizes, in all new forms, the simple data from instrumental measurements. There is, from all evidence, substitution of the metric noumena at the metric phenomena, and the thought that organizes the experiment is not at all a simple translation of the metric organization discovered in phenomena. The organizing thought had an entirely other inductive value. In these conditions, it would be reckless to set off from preliminary metric abstractions in order to fix a metaphysical border of the science of phenomena. This way, in our opinion, measurements are hardly achieved from furnishing the abstract and relative data that mathematical thought itself developed as a very apt thought to give schemas experimental concretization.

Are we able to draw some conclusions from the preceding remarks? It seems to us that these conclusions can be of two kinds: scientific and philosophic.

Scientifically, the border of knowledge appears to mark only a momentary arrest in thought. It would be difficult to trace objectively. It seems that the limitation of scientific thought is desirable whether it is in terms of the program or of the absolute obstacle, in terms of possibility rather than impossibility. One would wish that each science could be able to propose a kind of five-year plan.

Philosophically, every absolute border proposed in science is a mark of a problem badly posed. It is impossible to richly think impossibility. As soon as an epistemological border appears clear, it is then that it assumes the right to severe itself from the subject of first intuitions. But first intuitions are always intuitions to be rectified. When a method of scientific research loses its fertility, it is that the point of departure is too intuitive, too schematic; it is that the base of organization is too limited. The duty of scientific philosophy seems then very clear. It is necessary to wear away all parts of initial limitations, to reform non-scientific knowledge that still fetters scientific knowledge. Scientific philosophy needs in some manner to systematically destroy the bounds that traditional philosophy has imposed on science. It is in fear, in effect, that scientific thought guards not for the traces of philosophical limitations. In summary, scientific philosophy needs to be essentially a scientific pedagogy. So, in the new science, new pedagogy. That which we lack most is a doctrine for elementary knowing in agreement with scientific knowing. In brief the a priori of thought are not definitive. They also must be subjected to the transmutation of rational values. We come to realize the conditions sine qua non of the scientific experiment. We demand consequently that the scientific philosophy renounce the immediate real and that it aid science in its struggle against first intuitions. Oppressive borders are illusory borders.