Sciences, Philosophie, Histoire – UMR 7219, laboratoire SPHERE

:: How documents reflect research work as well as the social and material contexts of the circulation of works.

Organisation : K. Chemla (CNRS, SPHere)

• Carole Hofstetter (Paris 8-UPL)
  Reading circles of the Introduction to Arithmetic in Byzantium
  Nicomachus’ treatise Introduction to Arithmetic (1st-2nd century CE) has more than a hundred Greek manuscript witnesses, of which almost half were copied during the 13th and 14th centuries. The high degree of contamination in this part of the manuscript tradition makes it very difficult for classical philological methods to uncover and present the relationships between these manuscript witnesses in a precise manner.
  Through a series of case studies, I will show how a complementary approach, guided by a historical perspective aiming at identifying reading circles studying the text in which these manuscripts were produced and used in the medieval period, can prove relevant to counter this difficulty in the case of the Introduction to Arithmetic.

• Julia Kursell (Paris 8-UPL)
  Writing down music – Carl Stumpf and his Score of a Performance from Siam
  In 1901, philosopher and experimental psychologist Carl Stumpf (1848-1936) published a paper on the visit of a theatre group from Siam (now Thailand) to Berlin in 1900. During his work on musical perception and cognition up to then, he had become interested in phonographic recording, which he first rejected and then embraced. The recordings he made from the Thai musicians is nowadays considered to be the first in the Berliner Phonogramm-Archiv, which was to become the largest collection of ethnographic wax-cylinder recordings. This recording is special in that Stumpf tried to record the musicians separately to then recompose the performance as an edited score which he included in the paper. This edition will be held against other modes of writing he used in his work, such as short-hand and regular music notation.

• Costantino Moretti (École française d’Extrême-Orient)
  Random Notes on Standards, Textual and Formal Variants in Dunhuang Buddhist Manuscripts
  The notion of “mistake” and that of “textual variant” share, at times, some ambiguities. By comparing several manuscripts containing the same Chinese Buddhist scripture, for example, it becomes apparent that a number of presumed variants in modern edited texts are in fact simple scribal copying errors. In most cases, these mistakes are produced due to a chain of factors whose combination leads to accidental textual corruption, or due to a misunderstanding of the source-manuscript layout. In the first part of my talk, I will discuss presumed textual variants linked to possible scribal errors, with a focus on Dunhuang Buddhist material.
  Medieval catalogues of Buddhist scriptures provide details on the structure and/or internal organization of sūtras and commentaries in an early phase, also setting out interesting codicological details concerning book collections stored in official monastic libraries. By comparing early catalogues to later sources significant differences in the structure of various texts come to light, indicating that scriptures were, at times, circulating in several formats — possibly with a different layout — and that textual divisions may have evolved over the centuries. In the second part of my talk, I will point out additional considerations regarding variants/alterations that involve a modification of the text layout, i.e. “formal variants”, which can determine a misconstruction of the manuscript that served as the basis for producing a given copy.

November 10, 2022

:: Diagrams

Organisation : K. Chemla (SPHERE)

• Divna Manolova (Visiting Postdoctoral Fellow, Max Planck Institute for the History of Science)
  Scientific Illustrations in Late Byzantium: A Case Study of Eclipse Diagrams
  In this paper, I interrogate how diagrams operate as scientific illustrations in the material as well as social context of late Byzantine books and in particular, how eclipse diagrams (both solar and lunar) function as vehicles for knowledge transfer. Focusing on a sample of diagrams found in Byzantine cosmological and astronomical miscellanies from the thirteenth, fourteenth and fifteenth centuries, I discuss, first, what seems to constitute an eclipse diagram. To this end, I provide an overview of the main graphic and aesthetic conventions, on which an eclipse diagram relies in order to be understood. Second, I examine how an eclipse diagram encodes and organises knowledge about light and shadow, line and colour, as well as about relative distances and magnitudes involving the earth, the sun and the moon. Finally, I analyse how eclipse diagrams relate to corresponding narrative explanations of eclipse occurrences. My principal goal in this third part of the paper is to explore whether diagrams play a specific role and offer a contribution distinct from that of narrative in forming, structuring, preserving, and transmitting knowledge about astronomical phenomena (for the purposes of this paper, about eclipses) and whether by studying diagrams I could elucidate, ever so slightly, the ‘obscure’ place that is the so-called ‘late Byzantine classroom’.

• Matthieu Husson (SYRTE, Observatoire de Paris - Université PSL, CNRS, Sorbonne Université)
  A natively digital critic of astronomical diagrams in pre- and early modern astronomy: A short presentation of the EIDA project
  For centuries across Afro-eurasia the astral sciences were fostered to meet a wide range of religious, ritualistic and political needs, to observe, understand and predict the natural world. Consequently, in multiple interrelated cultural contexts, practitioners compiled manuscripts and early prints including numerical tables, texts and diagrams, of which tens of thousands are extant. Building upon the results of the TAMAS project dedicated to the collective elaboration of tools and publications establishing new standards for natively digital analysis and edition of astronomical tables the ANR project EIDA will focus on diagrams as a visual heritage and astronomical or mathematical tools of reasoning in the astral sciences.
  EIDA is a project which associates research in computer vision with history of astronomy in a deeply interdisciplinary perspective. The ambition of the project is thus to develop and implement a new approach in the study of astronomical diagrams in pre-modern and early modern periods by creating and leveraging novel and specific automatic image analysis tools. The two main objectives of EIDA are: (1) to shape a specific criticism of astronomical diagrams and standards for their natively critical edition addressing, through their material and epistemological aspects, the variety of their functions and circulation modes, and (2) to take this innovative analysis to an unprecedented Afro-Eurasian scale by developing new computer vision approaches, capable of decomposing an astronomical diagram into components that are meaningful for analysis and edition, without relying on dataset-specific human annotation.
  In the first part of the presentation, I will focus on the interdisciplinarity of EIDA and the challenges it induces for the history of science dimension of the project. In the second part of the presentation, I’m going to focus more specifically on the history of science corpus and issues we hope to explore in the project (doing my best to balance my personal bias toward Latin mediaeval sources).

• Alban da Silva (SPHere)
  A Mathematical Study of Diagrams on Vanuatu sand
  Il existe dans les sociétés traditionnelles du Vanuatu (Pacifique Sud) une activité culturelle consistant à dessiner, à même le sol, des figures symétriques à l’aide d’un doigt. Le dessinateur produit une ligne continue qui ne repasse pas continument sur elle-même, il ne lève pas le doigt durant le tracé et il finit son dessin au point de départ.
  Dans un premier temps, je décrirai comment une enquête ethnographique menée chez les Raga de l’île de Pentecôte m’a permis d’élaborer un modèle permettant de suggérer la dimension mathématique de cette pratique, notamment algorithmique. Je montrerai qu’il peut être fructueux de considérer les méthodes de tracé comme des chemins eulériens dans un graphe bien choisi. Il est alors possible de faire des hypothèses sur les processus qui conduisent à la création et à la mémorisation de ces dessins. Cette méthodologie pouvant constituer une étape vers la compréhension des idées mathématiques développées chez les Raga, elle pourra intéresser à ce titre les historiens cherchant à analyser des diagrammes utilisés dans différents contextes culturels et historiques.

December 8, 2022

:: Bibliographies

Organisation : K. Chemla

• Clément Bonvoisin (ED 623, Université Paris Cité, SPHere) (postponed)
  As if we worked in a vacuum:" making bibliographies on optimal control problems in Lev Pontryagin’s group

• Marie Lacomme (ED 623, Université Paris Cité, SPHere)
  Comment aborder une bibliographie en histoire des sciences ?
  En 1941, le biologiste Theodore Cedric Ruch publie Bibliographia primatologica, une bibliographie de trois-cents pages pensée pour rassembler la liste de toutes les publications scientifiques portant sur les primates non humains depuis le XVIIIᵉ siècle. L’ouvrage est dès sa parution salué par nombre de chercheurs, et fait l’objet de recensions élogieuses dans Nature, Science, et plusieurs autres revues académiques, anglophones, mais aussi francophones. Ces dernières années, plusieurs primatologues ont écrit que cette parution constituait un des éléments fondateurs de leur champ.
  Parce que le livre est important pour les acteurs, l’historien.ne des sciences travaillant sur l’histoire de la primatologie est amené.e à s’y intéresser. Mais comment aborder un texte essentiellement constitué de quatre-mille-quatre-vingt-dix références bibliographiques, et donc, difficilement lisible, en tout cas, in extenso ? C’est le problème que nous adresserons dans cette présentation, en exposant les diverses méthodes que nous avons envisagées et développées pour étudier la bibliographie de Ruch : examen du paratexte, de l’organisation de l’ouvrage, de la correspondance entourant la publication, mais également écriture d’un programme informatique pour analyser statistiquement son contenu.

• Edgar Lejeune (TEMOS & SPHere)
  Dresser un bilan, favoriser les contacts, légitimer des pratiques ? Fonctions d’une bibliographie dans un champ d’étude naissant, l’histoire médiévale numérique (1960-1981)
  En 1982, Caroline Bourlet, Carine Doutrelepont et Serge Lusignan publient une bibliographie intitulée Ordinateurs et études médiévales. Composé d’environ 1250 titres, ce travail s’inscrit dans la continuité de plusieurs autres initiatives qui ont pour objectif d’organiser, au sein des études médiévales, des façons de faire circuler les travaux et les méthodes des premiers utilisateurs des ordinateurs. Dans l’une des quelques recensions dédiées à l’ouvrage, René Pellen rend compte du fait que la bibliographie est appelée « à rendre les plus grands services en permettant de dresser des bilans » et « en favorisant les contacts » (Pellen, 1985). Dans une autre recension, Jean-Philippe Genet affirme que la bibliographie de Bourlet, Doutrelepont et Lusignan « prouve [...] que l’introduction de l’ordinateur dans les études médiévales a commencé à porter ses fruits » (Genet, 1983).
  Partant de ces premiers indices, nous proposons au cours de cette intervention de nous pencher sur les fonctions assurées par cette bibliographie dans le champ naissant des études médiévales assistées par ordinateurs au début des années 1980. Nous nous appuierons sur un entretien mené avec l’une des auteures, sur les archives du travail d’enquête à la base de la bibliographie ainsi que sur la bibliographie elle-même, et plus précisément sur son apparat critique (index-matière, index des noms, index des corpus).

January 12, 2023

:: Notations, symbolism

Organisation : K. Chemla & A. Remaki

• Karine Chemla (CNRS, SPHere)
  To put or not to put? That is the question
  In terms of mathematical texts, two types of sources came down to us from early imperial China: some are manuscripts excavated from tombs and others are canons handed down through the written tradition. Both types of text are as a rule composed of problems and procedures solving these problems. Moreover, these texts all refer to calculating rods being used on a material surface, on which computations were carried out. However, before the 10th century, none of these texts contain any illustration showing how rods were used in the context of computation. Some procedures explicitly refer to the operation of putting down numerals on the calculating surface, some do not. This feature seems to depend on the mathematical topic dealt, with and also to depend on the nature of the text in which the procedure is inserted. What can these differences tell us about the history of the calculating surface and its use? What can they tell us about the difference in nature between different types of sources? These are the questions that this talk will address.

• Arilès Remaki (Erc project Philiumm, CNRS, SPHere)
  Implementing term relations in two-dimensional space: the case of Leibniz’s table
  A table is a complex object. The question of whether it is relevant or even possible to reduce a table to a linear sequence of signs or instructions must be asked and allows for very important semiological reflections. Nevertheless, the different modalities of production and presentation of knowledge resulting from historical analysis and exegesis require us to find, in concreto, data structures within which to implement the substance of the material artefacts that are the tables. Thus, without ever ceasing to bear in mind the impossibility of capturing in a discrete formula the phenomenological complexity of such diagrams, the work of abstraction forces us to choose which aspects to keep, simplify or delete.
  The presentation will focus on the particular case of a digital edition project of Leibniz’s manuscripts, in which many tables are found. The question that will be discussed is how to implement in parallel the geometric aspects of the table and the combinatorial relations between the terms in a formal language (in this case: the TEI language).

• Mark Geller (UCL, Londres)
  Medicine at the ends of the Silk Road
  The discovery of Chinese medical, magical, and hemerological texts from Mawangdui has raised important comparative questions which have so far been hardly addressed. Using Donald Harper’s editio princeps as a guide, the present talk will attempt to remind us that Mawangdui texts are not only early and atypical for Chinese medicine, but that they also resemble the kind of recipe-based medicine best known from Mesopotamia. Is this coincidental, or could both ends of the Orient have shared a somewhat similar contemporaneous scientific culture? This paper intends to provoke questions, not answers.

February 9

:: Adaptations, Abridgments, Reformulations, Summaries

Organisation : A. Keller & Julie Lefebvre (Université Paris Nanterre, MoDyCo - UMR 7114)

Scholarly and scientific literature contains many texts that are self defined as adaptations, abbreviations, reformulations or summaries of either another canonical text or more generally traditional knowledge. These compositions have often been understudied or neglected, considered as secondary texts. In this seminar we take specific case studies to explore the textual and linguistic acts that the writing of such texts implies.
What are the projects by which authors set out to write these texts which aim at re-telling another text differently, perhaps in a shorter way?
What are the linguistic and textual acts, explicit or not, by which these summaries or reformulations are made?
What are the words used by our actors to evoke these actions? do they help us discriminate and think about what are reformulations, abbreviations, summaries or adaptations?

• Julie Lefebvre (Université Paris Nanterre, MoDyCo - UMR 7114) (CNRS, SPHere)
  The book review, an academic written practice. The case of Jean Starobinski’s Mots sous les mots (1971)/
  We will focus on the book review (or book report) as it is used in the academic sphere. The writings that result from this practice - the reviews themselves -, although considered secondary to the text that is the subject of the review, play a central role in the history and structuring of science (Sarton 1924, 1960). In this presentation, we will focus on these writings from a linguistic and discursive point of view, as they belong to a larger group of discourse genres whose function is to ’take the place’ of another discourse (Authier-Revuz and Lefebvre 2015).
  We will thus examine the characteristics of the book review in the humanities and social sciences by comparing it with the summary, another academic writing that "takes the place" of a text considered as primary; and by comparing different reviews of the same work. We will base our observations on the summary and the reviews of Jean Starobinski’s text devoted to Ferdinand de Saussure’s research on poetic anagrams,Les Mots sous les mots (Gallimard, Paris, 1971), a work which gave rise to various reviews at its publication (Testenoire and Willemin 2021).

• Agathe Keller (CNRS, SPHere & Université Paris Cité)
  Issues concerning reformulations in Sanskrit mathematical and astral texts: some examples
  In the preamble of Sanskrit astronomical treatises, the history of astral sciences is sometimes told as the history of multiple reformulations of what was taught by predecessors, sometimes divine. After having read texts which designate both texts and knowledge as attempts at re-formulation, we will examine two cases of reformulation of a different nature.
  First, we will look at standard verses from Sanskrit mathematical texts (notably the definition of the decimal place-value notation), looking at the different ways in which the sūtras of treatises have been composed, and what the comparison of texts tells us about the ways in which they have made reformulations.
  Then, we will examine a case of allusive paraphrase of a prose commentary in other prose commentaries, concerning the construction of armillary spheres, to discuss together if it seems to you to be a case of reformulation that does not say its name.

• Zeinab Karimian (The Written Heritage Research Institute, Tehran)
  The Summary of Conics: A Redaction of the Conics of Apollonius by Maḥmūd ibn Qāsim al-Iṣfahānī (12th century)
  The treatise of Conics of Apollonius (3rd-2nd BC) was translated into Arabic in 9th century. Since then, this translation provided a pathway to several areas of new research in this branch of mathematics. Moreover, several redactions of this treatise were composed in Arabic, in order to facilitate access to the Conics and to enrich the initial text. Maḥmūd ibn Qāsim al-Iṣfahānī, the 12th century mathematician, wrote his redaction, entitled The Summary of Conics, through a new formulation of the whole text, with applying new technical terms. In this talk, I will present this treatise and its role in history of mathematics. Afterwards, we will examine the methods of Iṣfahānī via some examples of his redaction, and we will see how Iṣfahānī reformulated the propositions of Apollonius in order to provide his own redaction of the Conics

March 9

:: Genesis

Organisation : F. Bretelle-Establet (CNRS, SPHere), M.-C. Bustamante et Emmylou Haffner (ITEM)

• Jan von Plato (University of Helsinki)
  Deciphering Gödel’s shorthand manuscripts
  Kurt Gödel (1906-1978) did practically all of his writing in an obsolete German shorthand. Such writing would usually be done by a court reporter or secretary who soon afterwards, partly based on fresh memory, would produce a typewritten text. Gödel published very little after 1940, but he left behind literally thousands of pages of scientific manuscripts in shorthand, mostly about logic and mathematics, but also on such diverse topics as physics, philosophy, and theology.
  Based on experience in the past five years on transcribing Gödel’s notebooks, the character of shorthand writing is illustrated, with an emphasis on special aspects of sources from logic and foundations of mathematics. Work on Gödel’s manuscripts has turned out a challenging but also rewarding task, in that the results deepen and change our understanding of the achievements of one of the icons of 20th century science. The study also profoundly affects previously established expertise on the topic as very few, a mere handful at present, master the script.

• Martha-Cecilia Bustamante (SPHERE)
  Taper à la machine un texte écrit à la main. Un exemple des enjeux autour de cette pratique
  Dans cet exposé nous allons présenter des documents, inédits, manuscrits et tapés à la machine de Jacques Solomon. Nous regardons, en premier lieu, les manières dont, les uns et les autres ont été composés. Nous verrons ensemble ce que la comparaison des textes donne sur le processus à la fois de création et d’écriture scientifiques. Du point de vue méthodologique et de l’histoire des textes, nous montrons l’intérêt de se porter sur l’étude des documents tapés à la machine. Dans notre cas, l’étude est d’autant plus pertinente que les documents portent sur un même sujet : la théorie quantique relativiste de Paul Dirac. Enfin, si dans un premier temps, ils montrent Jacques Solomon « en train de taper à la machine un document manuscrit, donc préparé à l’avance », dans un deuxième temps, ils montrent surtout les enjeux d’une pratique où la réflexion « se faisant par étapes », a lieu dans un temps long, voir indéfini.

• Clément Bonvoisin (ED 623, Université Paris Cité, SPHere)
  As if we worked in a vacuum:" making bibliographies on optimal control problems in Lev Pontryagin’s group
  From 1956 to 1961, a group of Soviet mathematicians, composed of Lev Pontryagin (1908 – 1988) and three of his former students, published a series of articles and a monograph on optimal control problems. These problems had emerged from discussions in certain scientific and engineering communities, in different countries, during World War II and the early Cold War. The general issue was to find the best ways to control a given system, in order to transfer it from its initial state to another, desired one. Through their works, Pontryagin’s group put forward a result they called the maximum principle, depicting it as a decisive tool to solve optimal control problems.
  In this talk, I will focus on one aspect of these published works: their bibliographies. As it occurs, these bibliographies were not comprehensive lists of the existing literature on optimal control problems known to Pontryagin’s group. Rather, they were the results of choices made by the authors of these texts. These selections of works therefore become objects for historical inquiry. Would the authors tailor their bibliographies in order to cast light on certain works and overshadow others? How does their choice relate to the reminiscences of one of the authors, Vladimir Boltyansky (1925 – 2019), according to whom ‘Pontryagin did not like to indicate scientific sources out of his group […], as if we worked in a vacuum without any connection with world science?’

April 13

:: Crossing out

Organisation : E. Haffner

• David Rabouin (CNRS, SPHere et ERC Philiumm)
  De l’importance et de la difficulté d’éditer les ratures
  Abstract: L’étude des ratures est un élément essentiel pour tenter de reconstituer le cheminement de pensée d’un auteur comme dans son acte même. Il s’agit de ce qu’on pourrait appeler l’approche locale des ratures. Mais il est également possible de trouver dans les ratures des éléments d’une évolution globale, des points où la pensée d’un auteur change pour évoluer vers quelque chose de nouveau. C’est notamment le cas quand on peut repérer des ratures/corrections que l’auteur fait systématiquement. Ce type de ratures n’a pas encore fait, à ma connaissance, l’objet d’une analyse méthodologique approfondie, alors même qu’elle est de très grande importance et pose des problèmes éditoriaux spécifiques. Dans cet exposé, je donnerai trois exemples de cette approche globale dans le cas des manuscrits mathématiques de Leibniz.

• E. Haffner (ITEM)
  Que sont les ratures et à quoi servent-elles ?
  Abstract: Les brouillons du mathématicien, qui nous donnent à voir les processus de recherche et d’écriture de leur auteur, contiennent notamment un certain nombre de ratures. Dans cet exposé je propose une première réflexion sur le statut et rôle des différents types de ratures dans les brouillons mathématiques. Je suggérerai qu’il existe plusieurs sortes de ratures – on ne corrige pas que les erreurs – et que celles-ci témoignent de différents actes d’écriture, d’un auteur à l’autre mais aussi chez un même auteur. Je me concentrerai sur des textes allemands et français du 19e et début du 20e s.

May 11

:: Note taking

Organisation : M.-C. Bustamante (SPHere) and K. Chemla (CNRS, SPHere)

• Sara Confalonieri (Université Paris Cité, HPS, SPHere)
  Fourier’s algebraical manuscripts: what the text can say about composition practices
  We will focus on a manuscript text of Fourier’s from 1804 which deals with the enumeration of the real roots of an algebraic equation. This text is particularly interesting because there are two handwritten versions (one by Fourier, the other by collaborators). Our aim will be to identify Fourier’s writing practice through the hints given by the materiality of the manuscript.

• Wang Xiaofei (IHNS, Beijing)
  How the notes from Fourier’s course at the Ecole Polytechnique were produced
  The manuscripts Ms. 1852 and Ms.2044 are two individuals’ notes deriving from Joseph Fourier’s course of analysis at the Ecole Polytechnique. Each manuscript contains the notes of the lectures Fourier gave to different classes of students at the school during the year 1796. This talk will focus on the process of the production of these notes. Through an analysis of the textual features of the two manuscripts, it aims to clarify their factual relationship with Fourier’s teaching.

• Nathalie Sage-Pranchère
  
  Ecrire la recherche médicale, entre preuve et transmission : l’exemple des érythroblastoses fœtales pendant la première moitié du XXe siècle
  La récente pandémie de covid-19 a soudainement mis en lumière dans l’espace public les pratiques d’écriture scientifique en médecine, la temporalité des publications, la place des méthodologies, le nombre souvent considérable des auteurs et dès lors, le statut multiple et complexe de l’autorité pour ces écrits. Ces pratiques sont toutefois relativement récentes à l’échelle de l’histoire de l’écrit médical et se structurent progressivement après la Seconde Guerre mondiale tandis que le modèle de l’evidence-based medicine, ce qu’Othmar Keel définit comme la « médecine des preuves », devient prédominant. À partir de l’exemple des érythroblastoses fœtales, un regroupement pathologique qui se constitue à l’orée des années 1920 à la frontière de plusieurs spécialités médicales, je m’intéresserai à la période qui précède celle du triomphe de l’evidence-based medicine. J’étudierai dans les publications périodiques médicales européennes et nord-américaines la manière dont se construisent et s’écrivent les savoirs sur cette maladie. La critique externe sera mobilisée pour analyser la forme des articles (organisation de l’argumentation, emplacement, présentation et importance des cas cliniques, recours à des illustrations, présence et présentation de la bibliographie), leur place à l’échelle des revues, leur fonctionnement en réseau avec d’autres publications du ou des mêmes auteurs. La critique interne permettra de déceler dans l’écriture les traces des actions et des objectifs du ou des médecins-chercheurs.
  
  
  Download program
  
  

  

June 8

:: Commentaries

Organisation : K. Chemla and A. Keller

• Marie Bizais-Lillig
  Kong Yingda’s subcommentary as a rhetoric device in Tang China scholarship
  
  Résumé: tba

• Guillaume Loizelet
  Kushyar’s (11th c.) comments on al-Battani (10th c.) : to reiterate, to explain, to essentialize.
  
  Résumé: Medieval bibliographers inform us that the practice of commentary was widespread among astronomers writing in Arabic in the tenth and eleventh centuries. However, few of these texts have come down to us.
  The presentation will focus on a short text on the distances and sizes of wandering stars written by Kushyar ibn Labban at the beginning of the tenth century, which turns out to be a commentary on a text written by al-Battani a century earlier: Kushyar does not mention al-Battani, but takes up his vocabulary, clarifies some of his predecessor’s excessively rapid passages, and pushes to the extreme the primacy accorded to observation that can be discerned in the background.

• Organisation for next year
  
  
  Download program
  
  

  

. . . .

VENUE

Building Condorcet, University of Paris, campus Diderot, 4, rue Elsa Morante, 75013 - Paris*. Plan.
Calculate your itinerary with RATP

Metro : lines 14 and RER C, stop: Bibliothèque François Mitterrand ou ligne 6, stop: Quai de la gare. Bus: 62 and 89 (stop: Bibliothèque rue Mann), 325 (stop: Watt), 64 (stop: Tolbiac-Bibliothèque François Mitterrand)