Dominique Raynaud

Ce texte est une exploration des manifestations de la « pensée partisane » aux marges de la science. La science répondant à une norme d’objectivité et l’objectivité étant incompatible avec l’esprit partisan, repérer la présence d’une pensée partisane en science revient à pointer une contradictio in adjecto. Par suite, les éléments partisans qui auront été identifiés devront être placés à la limite ou au-delà du domaine de la science. L’expression « pensée partisane » se trouvant dans le voisinage immédiat de l’idéologie, je rappellerai le sens de ces mots en indiquant sur quelle base on peut les distinguer. J’introduirai ensuite deux exemples – anthropocène et technique – qui me paraissent illustrer la pensée partisane dans le domaine des sciences et techniques. Par des voies très différentes, ces deux idées sont débitrices du contexte de la Guerre froide, ce qui n’est évidemment pas indifférent pour comprendre les mécanismes de leur formation et de leur diffusion.

This article studies the legacy in the West of Abū al-Wafā’s Book on those Geometric Constructions which are Necessary for Craftsmen. Although two-thirds of the geometric constructions in the text also appear in Renaissance works, a joint analysis of original solutions, diagram lettering and probability leads to a robust finding of independent discovery. The analysis shows that there is little chance that the similarities between the contents of Abū al-Wafā’s Book and the works of Tartaglia, Marolois and Schwenter owe anything to historical transmission. The commentary written by Kamāl al-Dīn Ibn Yūnus seems to have had no Latin legacy either.

The present paper discusses some new findings on the topic of Masaccio's Trinity fresco perspective reconstruction. Some scholars have tried to reduce the fresco's anomalies with the help of photogrammetrical reproductions and computer analysis. On this basis, it has been argued that Masaccio used the normal case of costruzione legittima. It is very likely that Masaccio took the lines of the plane joining the abaci of the capitals as a guide for drawing the vanishing lines of the coffered vault. However, in doing so, he followed an erroneous method of construction disregarded till now. The aberrations thus discovered strongly support the view that Masaccio's Trinity fresco is not a model of linear perspective. The thesis that Masaccio designed the fresco with the help of a ground plan and elevations is dubious, and the search for the distance point is condemned to remain an unending problem.

The study of the dispute that broke out between the medical schools of Paris and Montpellier (1817-1852) shows that professional interests, but also philosophical and political factors drew the fitgets of Montpellier's school about its doctrine. But these factors are inable to explain the scientific contents. Further, the closure of the controversy did not result from a social negotiation of truth. It simply favoured the scholars who had major scientific productivity. This dispute so contradicts the principles of the relativistic program in the field of the sociology of scientific knowledge, and shows that some studies could have misunderstood the sense of a social determination of scientific contents. L'analyse de la controverse qui éclata entre les écoles de médecine de Paris et de Montpellier (1817- 1852) montre que des intérêts professionnels, mais aussi des facteurs philosophiques et politiques conditionnèrent la crispation de l'école de Montpellier sur ses positions doctrinales. Mais ces facteurs n'expliquent pas le contenu des théories. Le règlement de la controverse fut étranger à une négociation sociale de la vérité, et tourna à la faveur ceux qui eurent la productivité scientifique la plus élevée. Cette controverse, qui échappe aux principes du programme relativiste en sociologie des sciences montre que certaines études peuvent s'être méprises sur le sens d'une détermina- tion sociale des contenus scientifiques.

This article aims at clarifying sociological definitions with the help of an original tool, the logical table of contingency, which has characteristics in common with both the statistical table of contingency and Carroll’s bilateral diagram. The unclear notion of ghetto is taken as a test-case. Successive tables of contingency are applied to various defining properties and various sociohistorical situations. Cladistic methods are then used to make the relationships between all ghetto-like situations explicit.

Est-il vrai que la science et la technologie forment un tout ? Qu’elles sont unies par des liens constants et nécessaires ? Est-il vrai que le raisonnement technologique prend toujours place dans un contexte industriel ? Que les bio- et nanotechnologies forment la part essentielle des technologies contemporaines ? Que l’application définit adéquatement la relation entre la science et la technologie ? Dans ce livre, Dominique Raynaud examine les limites de chacune de ces idées. Il dresse un panorama complet des problèmes que pose la technologie au moyen d’une étude détaillée des aspects cognitifs, ergologiques, professionnels et juridiques de la relation entre la science et la technologie et met à nu la fiction utilitariste de la prétendue « technoscience ». Tout aussi éloigné de l’apologie que de la critique radicale, ce livre défend une perspective empirique et analytique dans l’étude de la technologie.

The idea that linear perspective arose only in the West due to the strength of an unusual process of rationalization is denied by the fact that IXth century Islamic scholars had yet a thorough knowledge of the optical and geometrical materials required in perspective. In addition, the process of rationalization was rarely so intense as in that time, because truth uniqueness and scientific communalism were core values of Medieval Islam. The puzzle is not a matter of less or more rationality, but a matter of axiological motives. First, Islamic disregard for perspective is the result of a value-set insofar painters were fearing the hadīths' prohibition, or wished to avoid pretentiousness of their name (muṣawwir). Second, Islamic art did not draw from Qurʿān and hadīth's precepts directly. Third, painters implemented axiological motives, to put iconic practice in accordance with aniconic leanings.

This article explores how “partisan thinking” presents itself in the borderlands of science. As science conforms to an objectivity norm, and objectivity is incompatible with the partisan mindset, to discover a latent partisan notion in scientific discourse is to lay bare a contradictio in adjecto. It follows that any partisan elements identified should be placed on the edges of, or even outside, the domain of science. The term “partisan thinking” is a close neighbor to ideology, and so I will briefly review the meaning of each word and explain how they might be distinguished. I will then introduce two examples – the Anthropocene and the French concept of technique – that illustrate the presence of partisan thinking in the realm of science and technology. By two very different routes, each of these ideas carries echoes of the Cold War, and this is clearly of interest to us in seeking to understand how they came to be formed and to pass into popular usage.

L'étude de la conception architecturale se trouve au carrefour de trois approches: sciences de la conception, psychologie cognitive et architecturologie. Cette dernière tente de modéliser toute la gamme des changements d'état qui apparaissent durant le processus de conception. L'architecturologie sait décrire, au moyen d'échelles, les opérations de conception par lesquelles l'architecte attribue des mesures à l'édifice. Cette modélisation paraît adéquate lorsque le modèle morphologique n'est pas transformé durant le processus de conception (comme dans certains projets de Jørn Utzon ou de Alvar Aalto). Dans d'autres cas – comme dans certains projets de Philibert de l'Orme, Ludwig Mies van der Rohe, Norman Foster – les changements d'état altèrent la structure même du modèle morphologique. De telles transformations ne peuvent plus être décrites par l'échelle. Reprenant une idée de Herbert A. Simon, l'article suggère que la description de ces nouvelles transformations appelle une autre classe d'opérateurs. Ce pourrait être des schèmes, si l'on entend par ce terme la classe des actions prototypiques qui peuvent être exprimées par des verbes fondamentaux comme ouvrir, fermer, séparer, lier, etc.

In this chapter, I discuss the "common mind thesis" that social integrates have mind of their own – a thesis which is built on a kind of aggregation paradox named doctrinal paradox. External analysis objects its very name: there is no paradox, but non-commutativity of the function composition involved in the aggregation process (Part I). Internal analysis shows that the doctrinal paradox works only when premises are separable. However premises turn out to be hardly separable in the examples cited in the literature (Part II). As the doctrinal paradox is sometimes considered as an argument of holism, its flaws questions the thesis that groups have minds of their own.

This paper examines the question of error in perspective from the viewpoint of the painter, not the spectator. This distinction significantly modifies the way in which perspective is approached for one must view it with the eye of the painter or the architect who constructed it. The perspective is therefore judged in terms of the methods used by its creator, which is to say in terms of the goals he set himself and the means at his disposal in order to achieve them. We then explain the "good reasons" the Renaissance painters had to consent to the three main types of error in perspective: "accidental errors" (type I), "ad hoc errors" (type II) and "systematic errors" (type III).

This article aims at contributing to the methods of the sociology of science, from an empirical study of the determinations of the velocity of light between 1676 and 1983. Far from being constructed and deconstructed at pleasure, the values of c have undergone a tendentially unidirectional and irreversible process of revision. The competing methods remained in the running as long as they produced an uncertainty less than, or equal to, the best known value. The analysis of “entrances” and “exits” of the methods on the research front makes clear the nature of scientific competition, which is a form of “redistributive emulation” that owes nothing to political rivalry or economic competition. By achieving results that differ from constructivism on the themes of literary inscription, games of construction/deconstruction, negotiation, and persuasion, this article outlines a kind of internalist sociology of science, which aims at exploring all aspects that make science science.

From Panofsky, it is usually considered that vanishing axis compositions result from the rules of curvilinear perspective. This article disproves this evidence. It is shown that a lot of vanishing axis compositions can find support on the principles of binocular vision (hyp. I) exposed in medieval optics, rather than on any type of linear (hyp. II) or curvilinear perspective (hyp. III). The comparison of these three hypotheses shows that only the first one fits well with a set of medieval and renaissance paintings, while avoiding inconsistencies arising – and not yet satisfactorily explained – in the framework of linear or curvilinear perspective. Depuis les travaux de Panofsky, il est habituellement considéré que les compositions en axe de fuite résultent d’une application des règles de la perspective curviligne. Cet article réfute cette thèse. Il est montré qu’un grand nombre de compositions en axe de fuite peuvent s’expliquer par les principes de la vision binoculaire que l’on trouve dans les traités d’optique médiévale (hyp. 1), plutôt que par quelque autre type de perspective linéaire (hyp. 2) ou curviligne (hyp. 3). La comparaison de ces trois hypothèses montre que seule la première peut rendre compte d’un ensemble de peintures du Moyen-Âge et de la Renaissance, tout en évitant les incohérences qui surgissent – et ne sont pas adéquatement expliquées – dans le cadre de la perspective linéaire ou curviligne.

Ruggero Bacone (ca. 1220–ca. 1292) nacque a Ilchester, nel Sommersetshire, in una data sconosciuta intorno al 1214, o più probabilmente 1220. Bacone ha studiato a Oxford (1234–42) prima di andare a Parigi, dove diviene maestro d’arte nel 1242 ed insegna subito dopo la filosofia di Aristotele. Ritornato a Oxford tra il 1248 e il 1251, abbandona la filosofia di «studio e commento» per dedicarsi allo studio della scienza sperimentale – una svolta decisiva che potrebbe essere spiegata dalla sua introduzione nell’ambiente di Roberto Grossatesta e del sue discepolo Adam de Marisco, magister regens dello studium di Oxford…

Dans les Due regole della prospettiva pratica, Vignola et Danti condamnent un système hétérodoxe de perspective à deux points de fuite. Après avoir analysé les épures de trente tableaux du Moyen Âge répondant à cette définition, il est montré que ces oeuvres ne répondent à aucun type connu de perspective. Elles échappent notamment aux règles de la « perspective synthétique » (Panofsky, White) et à celles de la « perspective bifocale » (Parronchi). Des arguments sont alors avancés pour reconnaître une application des principes de la vision binoculaire contenus dans les traités d’optique médiévale (Ibn al-Haytham et ses commentateurs latins). Sous cette hypothèse, la restitution des géométraux à partir des vues perspectives conduit à des résultats cohérents et conformes aux modèles architecturaux contemporains. Ces arguments incitent à reconnaître une application qualitative de la « perspective centrale binoculaire » au Moyen Âge.

Shadows have long been a central topic in the study of optics. The distinction between partial shadow (or the penumbra) and full shadow has been theorized by Maurolico and Kepler. Shadows are also dealt with in architecture and perspective, because they are directly involved in the rendering of volume and depth. Given his research interests, the subject of shadows was bound to capture the attention of Leonardo da Vinci, and it is known that he paved the way to the tracts on shadows that were appended to perspective treatises from the sixteenh century on…

L’article étudie l’inflation du nombre de coauteurs qui touche aujourd’hui plusieurs disciplines comme la physique des particules et les sciences biomédicales. Approfondissant l’exemple des collaborations ATLAS et CMS du CERN qui ont découvert expérimentalement le boson de Higgs en 2012, l’article précise les modalités de signature dans ce champ, étudie les facteurs responsables de l’inflation du nombre de signataires, se demande si cette inflation a provoqué l’apparition de phénomènes inédits, caractérise ces nouveautés, et précise le seuil temporel et numérique auquel elles sont apparues. L’article conclut que l’inflation du nombre d’auteurs en physique des particules est un phénomène émergent (non additif), visible notamment à l’activité normative engagée au sein de ce champ.

Si presume che Erazm Ciołek Witelo (Vitelion, Vitellio, Vitello, Vitulio, ecc., ca. 1230–dopo 1278) sia nato a Legnica, nella bassa Silesia, da padre tedesco e madre polacca, come suggerito dalla dedica dei Perspectivae libri decem a Guglielmo di Moerbeke (Veritatis amatori fratri Guilielmo de Morbeka, Witelo filius thuringorum et polonorum eterne…) Witelo si laureò presso la facoltà degli arti di Parigi negli anni 1250, prima di studiare diritto canonico a Padova tra il 1262 e il 1268. Nel 1268–9, si unì alla corte di Viterbo, dove fece amicizia con Guglielmo di Moerbeke, traduttore di varie opere dal greco al latino, come quelle di Aristotele o Proclo e, per quanti riguarda l’ottica, la Cattotrica di Erone di Alessandria (Vaticano, BAV, Ottob. lat. 1850)…

This article aims at clarifying what European scholars of the Middle Ages and the Renaissance knew about al-Kindī’s optics. Among the issues which deserve attention is the legacy of the Liber Jacob Al-Kindi de causis diversitatum aspectus. The application of the tracer method – a refinement of the analysis of text parallels – establishes that Latin perspectivists knew and exploited al-Kindī’s work: Bacon and Pecham by direct borrowing, Witelo by paraphrase. However, contrary to Werner, Björnbo, Vogl, Lindberg and Federici Vescovini’s opinion, Leonardo da Vinci had no access to al-Kindī’s work, which he knew only through some fragments replicated by Bacon, Pecham and, more incidentally, Witelo.

Most studies on scientific sources are conducted using the text parallels method, which consists in presenting two texts opposite each other in order to establish or exclude a correspondence between a «source text» and a «target text.» We have explained elsewhere the limits of this method, especially when (exact) text parallels turn into (approximate) doctrinal parallels, and when the potential sources are numerous. In this context, the search for sources is doomed to failure. In order to operate in this difficult environment, we therefore propose to enhance this method with a special focus on words with low lexical frequency, called «tracers,» in order to better detect their offspring in the target…

This paper examines the hypothesis that early perspective paintings were drawn arithmetically, without vanishing points. The best argument for this hypothesis is that the division of two parallel lines by straight lines intersecting each other at the vanishing point (geometrical method) is equivalent to the division of those parallel lines in proportional parts (arithmetical method). If arithmetical method had been used, then the vanishing points exhibited ex post should be purely fortuitous. But the lack of multiples and submultiples of measurement units, the absence of proportionality ratios, the length of the operating series, and the correspondence of vanishing points to visible loci of the picture offer sound objections for this hypothesis. The use of optics and geometrical method is more probative – though it does not mean that painters were using concepts of linear perspective, which would be an anachronism.

La “prospettiva”, come sistema di rappresentazione, deriva il suo nome dalla parola perspectiva che designava l’ottica nel Medioevo latino. Questo collegamento dipende dal fatto che alcuni dei princìpi della prospettiva provengono dalle teorie della visione. Tra i due insiemi, esistono sia rapporti di continuità che di discontinuità. La continuità è stabilita dallo studio dei paralleli testuali. Tuttavia esistono due biforcazioni principali che hanno segnato l’evoluzione successiva della perspectiva artificialis. Oltre il rapporto fra scienza e tecnica che le caratterizza, la perspectiva medievale era tricefala (optica, catoptrica, dioptrica), mentre la prospettiva sarà uno studio della visione diretta. La perspectiva presupponeva la visione binoculare, mentre la prospettiva linéaire adotterà la visione monoculare.

Poco si sa circa la vita di Abū Yūsuf Yaʿqūb ibn Isḥāq al-Kindī, conosciuto nel mondo latino come Alkindi. Secondo i bio-bibliografi arabo, sarebbe nato agli inizi del IX secolo a Kūfa, dove il padre serviva come governatore. Alcuni storici hanno sostenuto che al-Kindī avrebbe preso parte alla grande impresa di traduzione dal greco delle opere scientifiche guidata dal califfo al-Maʾmūn (regno tra 813-833) nella famosa Casa della Sapienza (bayt al-ḥikma), ma il bio-bibliografo più vicino alle fonti, Ibn al-Nadīm, inserisce al-Kindī dopo Tolomeo e Teone d’Alessandria, mentre i nomi dei sessanta traduttori della Casa della Sapienza appaiono in un’altra sezione del Fihrist.

Nous étudions les méthodes de construction des polygones réguliers proposées par quatre auteurs du XVII siècle : deux ingénieurs, Ambroise Bachot et Antoine de Ville, un érudit, Scaliger, et un mathématicien, François Viète. Certaines de ces méthodes se distinguent de la recherche de solutions exactes, typiques de la géométrie pure. Elles visent des solutions générales et approchées qui s’avèrent plus utiles aux ingénieurs, moins sensibles à l’exactitude mathématique qu’à des méthodes faciles à mémoriser et rapides à mettre en œuvre. La comparaison montre que les solutions de l’ingénieur et du mathématicien diffèrent quant aux connaissances mobilisées, quant au sens donné à la notion d’approximation et quant à ce que constitue une approximation acceptable. L’art des fortifications a impulsé une nouvelle approche des méthodes de construction des polygones réguliers. Cette approche échappe au canon des mathématiques pures non par ignorance, mais en connaissance des erreurs tolérées dans les constructions géométriques.

The study of architectural design bridges various approaches focusing on design sciences, cognitive psychology and architecturology. Architecturology aims to describe the full range of changes that appear during the design process. We are able to describe, by means of scales, the operations by which an architect gives size to the building. This modelling seems adequate when the morphological model is not adulterated during the design process (as in some projects of Jørn Utzon or Alvar Aalto). In other cases – as in some projects by Philibert de l'Orme, Ludwig Mies van der Rohe, Norman Foster – changes of state adulterate the very pattern of the morphological model. Such changes can no longer be covered by scale operators. Taking up an idea by Herbert A. Simon, it is suggested that the description of these new transformations requires another class of operators. These could be called schemata, with the term understood as the prototypical actions that could be rendered by basic verbs as open, close, separate, bind, etc.

Panofsky and White hypothezised that « axial constructions » result from a projective system in which object’s measurements are taken on an projection circle, before being moved to the picture plane (i.e. synthetic perspective). Although this construction has been yet criticized regarding painter’s ability to apply it, the idea of using synthetic perspective regularly reappears for it never has been examined in detail. But from a mathematical point of view, this system is simply impossible because, in a synthetic perspective, three vanishing lines taken on the same side of the axis cannot be parallel nor admit a convergence point. The falsification of Panofsky-White’s conjecture puts and end to a long series of contradictory judgements on this problem, and opens the way to a radical reinterpretation of axial constructions.