SCRIPT AS IMAGE: VISUAL ACUITY IN THE SCRIPT OF POGGIO BRACCIOLINI

The fact that the graphic substance of writing oscillates between text and image is a potential which writing carries in itself from the very beginning. Every graphic trace on the manuscript page relates to the conventions of time in a way that is determined by the scribe. This becomes particularly tangible when the conventions are deliberately and systematically broken and replaced by new ones on the basis of a concrete concept. By introducing the humanistic minuscule, a script developed on the basis of the historical model of the Carolingian minuscule, Poggio Bracciolini and his mentors and friends Coluccio Salutati and Niccolò Niccoli, created philologically revised copies of the texts of classical authors in what they called littera antiqua, the new old script. This paper wants to show how the conscious incorporation of elements of historical manuscripts and their transformation into a specifically humanistic product makes use of the graphical potential of script and mise-en-page in order to translate a humanistic discourse into SchriftBild.

The question of the materiality and visuality of books, beyond the concern with the texts they contain, arises in connection with Poggio Bracciolini's early work. For even before he could make his important manuscript discoveries during the Council of Constance (1414-1418), before he wrote his own literary works (his earliest work De avaritia was written between 1428 and 1429), before he followed in the footsteps of his mentor Coluccio Salutati (1331-1406, chancellor 1375-1406 and his friend Leonardo Bruni (1369-1444, chancellor 1427-1444 as chancellor of Florence (1453Florence ( -1458, he decisively shaped the appearance of the humanist book. «There was no humanism without books», writes Martin Davies on Italian Renaissance Humanism. He specifies: «[Books] were the prime material on which the movement was founded and the natural medium through which it was transmitted» (Davies, 1996: 47). Especially for the humanists Poggio, Salutati, and Niccoli, all of whom were particularly interested in grammar, the work on the book seems to have been, as per philological research tradition, above all work on the text, which they read, studied, corrected and reconstructed (Gombrich, 1976).
But their activity in relation to the manuscript as an object -the book itself -can be described primarily as searching, collecting, copying, and disseminating 1 . On the one hand, the manuscript represents the vessel in which the humanists find the text, and which they seek to free from the traces of its transmission by removing elements of its presentation, transmission faults, and other elements that belong to the ancient original. On the other hand, however, they transmit their restored text in the same vessel, because the revised text is finally presented to the public as a manuscript.
The script on the individual pages of the manuscript -developed by these early humanists, using older models and trials of Petrarch and others 2 -and its interaction with other elements of the mise-en-page such as decorated initials, hierarchies of different scripts, and letters or even page margins, become a medium for the self-presentation of a humanistic consciousness inscribed in the reproduction of the revised texts and thus a visual paratext on the ancient authors 3 .

The Practice of Copying
The philological work of Poggio Bracciolini in particular is wellknown today. The expertise that the Florentine humanist gained in this 1 To name only some examples, see: Greenblatt, 2011;Stadter, 1984;Flores, 1980. 2 Research on the development of the humanistic script and its dissemination has been ever-growing since the first major work by Berthold L. Ullman, The Origin and Development of humanistic Script (1974). Some of the most influential publications are: Ricci, 2016;Black, et al. 2016;De Robertis, 2006;Autenrieth & Eigler, 1988;Derolez, 1984;de la Mare, 1973 and1977. 3 «But this text is rarely presented in an unadorned state, unreinforced and unaccompanied by a certain number of verbal or other productions, such as an author's name, a title, a preface, illustrations. And although we do not always know whether these productions are to be regarded as belonging to the text, in any case they surround it and extend it, precisely in order to present it, in the usual sense of this verb but also in the strongest sense: to make present, to ensure the text's presence in the world, its "reception" and consumption in the form (nowadays, at least) of a book» (Genette, 1997: 1). While the English term «paratext» emphasizes its similar nature to the text, Genette's original French term seuil emphasizes the function of a threshold that shapes the reader's access to the text. Especially in the field of art history, and even more in the field of medieval book art, the term is often used in an extended form for non-textual but visual elements that consciously or unconsciously influence the act of reading and understanding the text. The use of the term paratext for the here described concept is more complex and includes other concepts as the iconicity of script and the idea of script as image. On these concepts see Hamburger, 2011 andMersmann, 2015;Merveldt, 2008: esp. 191-95. Contrary positions are formulated by Rockenberger & Röcken, 2009. See also Smith & Wilson, 2011 field was based above all on extensive study: the comparison of different sources and genres, but also of different versions of the same text, led to a conscious critical study of the works of ancient authors in particular. In his philological research Peter Lebrecht Schmidt makes an explicit appeal to Poggio's practice of comparative copying, in which Poggio not only consulted multiple versions of the same text, but also compared them side by side, compiling them in a revised version 4 .
While the philological aspect of this practice has had a strong reception, the material evidence of this procedure has so far been largely ignored. The visual effect of juxtaposing different manuscripts is almost obvious. For instance, for the version of Cicero's De legibus copied in Vatican City, Biblioteca Apostolica Vaticana, ms. Vat. Lat. 3245 (Fig. 1), Poggio compared at least two versions of the text: the ms. San Marco 257 ( Fig. 2) 5 , a Carolingian prototype that was written in the Abbey of Corbie in the 9th century and was probably brought to Florence by Poggio himself 6 , and ms. Strozzi 1066 (Florence,Biblioteca Nazionale Centrale,Magliabechiano XXIX,199), a copy on paper made in the 14th century in a partially cursive Gothic hand that was among Salutati's possessions and probably written especially for him (Fig. 3) Visually juxtaposing these two examples while comparing them with Poggio's version makes clear how the manuscripts functioned for him as both prototype texts and prototype images. While the Carolingian manuscript San Marco 257 has the clear writing that Poggio adopted for the design of the humanistic minuscule, he did not reproduce the two-column layout or the full-page decoration on fol. 1r (Fig. 4), the decorated initial of fol. 51v or the script hierarchy, which, in addition to the minuscule for the continuous text, uses a script tending towards 4 «Konkret dürfte der Bearbeitungsprozess so vorzustellen sein, daß Poggio F [San Marco 257] kopierte und dabei die zu kopierenden Partien laufend an einer danebenliegenden p- Handschrift [Naz. Magl. XXiX,199] kontrollierte, die von dort einleuchtenden Alternativen aber nicht nach F übertrug und von dort abschrieb, sondern direkt übernahm oder doch als marginale oder interlineare Varianten vermerkte», Schmidt, 1974: 282-283. 5 Florence, Biblioteca Medicea Laurenziana, ms. San Marco 257. The codicological details are described by: Ganz, 1990: 62, 154. See also Schmidt, 1974: 121f. 6 While Albinia de la Mare dated the manuscript between 1410 and 1415, Poggio rejoined the council of Constance only in 1414. It was only in 1415 that he searched the nearby monastery libraries for old manuscripts. Schmidt thinks that San Marco 257 was brought to Florence from one of the trips between 1415 and 1418. See de la Mare, 1973: 78n15;Schmidt, 1974: 122;Foffano, 1969. 7 Florence, Biblioteca Nazionale Centrale Firenze, ms. Magliabechiano XXIX, 199, (Strozzi 1066); see Schmidt, 1974: 238f. Schmidt suspects that the copy for Salutati is based on a corrected copy by Petrarch, which was lost in the copying process; see Schmidt 1974: 244-45. For the very large library of Coluccio Salutati, see De Robertis, et al. 2008;Ullman 1963. a Capitalis quadrata for the largest letters on fol. 1r, followed by an uncial script going down one level in the text ordering and a notch smaller in script size, a Capitalis rustica, a level further. Poggio's copy uses only a simple capitalis for the titles of the individual books. In addition, the manuscript has the single columns similar to the copy from the 15th century. But his copy contrasts with the cursive script used here, or more accurately, the different cursive hands and his minuscule produces a less hurried and calmer mise-en-page 8 . The contemporary model Strozzi 1066 is clearly an intermediate copy, shown by the use of partially cursive handwriting, the parallel work of eleven different scribes, and the use of paper, because paper manuscripts in Salutati's circle were basically transitional copies that served as models for more durable and representative copies on parchment 9 .
The version produced by Poggio is thus neither a pure copy of the manuscript's text nor of its script and form. Rather, Poggio actively interpreted the text on multiple levels when producing his revised version. Poggio's method of copying allows us to view the impressive libraries of Salutati and Niccoli that he had access to, not only as collections of texts from multiple historical epochs, but also as collections that document the visual and material aspects of these epochs' manuscript cultures 10 .

humanists' Collections -Books, Objects and Visual interests
The material aspect of the interest in collecting books is already evident in the earliest humanist collections: books, objects, and even contemporary works of art are repeatedly brought together in a single collection (Weiss, 1973: 59ff ). Niccolò Niccoli was known for his abundant collection of books and all sorts of ancient objects, as the artist Lorenzo Ghiberti describes:     'umanesimo,Mandragora,Firenze: 309,cat. no. 98;Cicero,De legibus,14th century, multiple scribes in gothic hands, partially cursive.
Besides other wonderful things I have had the pleasure of admiring in my lifetime, I would like to mention a wonderfully cut chalcedony that was in the possession of one of our fellow citizens, Niccolò Niccoli. He was a very ambitious man, researcher and collector of innumerable outstanding objects of antiquity, of writings as well as of Latin and Greek books. Among other objects of classical antiquity, he possessed this chalcedony, one of the most perfectly formed pieces I have ever seen […] 11 . The works collected by Poggio are less well-known, though «they were of sufficient quality to "delight a good artist"» 12 , as he proudly claimed. The juxtaposition of these objects -books and ancient artifacts -which are kept and used in the separate spheres of art collections and libraries today, raises the question of the extent to which their natures were not so clearly separated for humanist collectors. While the objects seem to arouse primarily visual interest, books are often seen only as documents and texts without taking into account their materiality and historicity as objects that have been transmitted over the centuries.
The humanists seem at first glance to differentiate in a similar way: Petrarch, in an attempt to relativize the joy of ownership and its proximity to greed (avaritia) 13 , makes a distinction between the joy of the ownership of objects and the collection of books. He stresses that «books please inwardly; they speak with us, advise us and join us together with a certain living and penetrating intimacy» 14 . The textual content stands above the object that one can possess materially. At the same time, he contradicts himself, as he clearly enjoys the manuscript of Homer's works sent to him by Nicola Sigero around 1354 -and which he can only possess since the copy was in the Greek original that he could not read: 11 «Fra l'altre egregie cose io vidi mai è uno calcidonio intaglio incauo mirabilmente et quale era nelle mani d'uno nostro cittadino, era il suo nome Nicholaio Nicholi: fu huomo diligentissimo et ne' nostri tempi fu inuestigatore et cercatore di moltissime et egregie cose antiche sì in scripture sì in uolumi di libri greci et latini, et infra' ll'altre cose antiche aueua questo calcidonio el quale è perfettissimo più che cosa io uedessi mai […]», cited from Bergdolt, 1988: 32-34. 12 «[…] effectus sum admodum capitosus […] habeo cubiculum refertum capitibus marmoreis, inter quae unum est elegans, integrum; alia truncis naribus, sed quae bonum artificem delectent. His et nonnullis signis, quae procuro, ornare volo academiam meam Valdarninam, quo in loco quiescere animus est; si tamen quies aliqua haberi potest in hoc procelloso mari», Bracciolini 1832: 214 (Bk. III, Ep. XVI). Translation cited from Thornton, 1997: 35. 13 «Although Petrarch often excused his own desire for books and, to a lesser degree, art as a sacred rather than a secular passion -"I flatter myself that the desire for noble things is not dishonorable", he wrote to the prior of San Marco, Giovanni dell'Incisa, around 1364 -he could not contain his lust for things», Findlen, 1998: 92. 14 «[…] libri medullitus delectant, colloquuntur, consulunt et viva quadam nobis atque arguta familiaritate iunguntur», Francesco Petrarca, Rerum familiarum libri i-Viii, 157, Fam. III, 18. See also Findlen, 1998: 92. Alas! Your Homer has no voice for me, or rather I have no ears for him! Yet the mere sight of him rejoices me, and I often embrace him and sighing over him I say: «O great man, how much I wish I could hear you!» 15 The admired object -the book -thus also belongs to Petrarch's collection along with other objects and shares their nature as artifact. It represents a bridge spanning over time to the past in which it was created, and thus, in addition to the information that can be derived from it as a document, it also represents a key to dialogue in its active dimension. For it must be emphasized that the collecting humanists were interacting productively with these early collections from the beginning. The bookseller Vespasiano da Bisticci describes Niccoli using daily the ancient objects he collected: [A]lways dressed in the most beautiful red cloth, which reached to the ground […] he was the neatest of men […] at the table he ate from the finest antique dishes […] his drinking cup was of crystal […] to see him at the table like this, looking like a figure from the ancient world, was a noble sight indeed (da Bisticci, 1995: 354).
In addition to this way of using the objects to bring antiquity to life, the artifacts were also part of a contemporary production of humanistic knowledge: art objects and manuscripts were seen as historical artifacts at the same time as they were used to produce new contexts. All objects in the collection were part of an active debate on content and materiality: a first example of this type is the historia imperialis (Fig. 5) by the Veronese Giovanni de Matociis, known as Giovanni Mansionario (d. 1337), begun around 1310. In his history of the Roman emperors, he not only uses the documentary information available to him from Roman coins, he combines the textual part of the historical work with the artifactual presence of the coin portraits by integrating them visually next to the text 16 . He thus translates not only the content but also the visual form into a humanistic product. In Petrarch's work as well, this visual side of humanist interest can be seen in the drawings of the busts of the authors he is currently reading. Wolf-Dietrich Löhr sees this insertion of portraits alongside the texts as an attempt to physically visualize the author: 15 «Homerus tuus apud me mutus, imo vero ego apud illum surdus sum. Gaudeo tamen vel aspectu solo et sepe illum amplexus ac suspirans dico: "O magne vir, quam cupide te audirem!"» Francesco Petrarca, Rerum familiarum libri XVii-XXiV, 45-46, Fam. XVIII, 2. Cited with the translation changes by Damen, 2012: 18. 16 The coins are not presented in a documentary form: instead Mansionario formally adopts the round format with profile portrait and transcription, but translates the ancient capitalis into a contemporary form. See Schmitt, 1974: esp. 190. Once again it becomes apparent that the abstract, immaterial character of the texts is not enough to worship an author, nor is the all-too-objective book. What was researched for is a physical image that can match the affect created by reading 17 .
In Petrarch's drawings as well, the artifacts, traditional portrait busts, evoke models even when there is no sculpture, as in the case of the drawing of a bust of Claudianus (Fig. 6) 18 .
Within this framework, the books in the humanist collections must also be seen in terms of their artifactual nature, i.e. as handed down historical objects that build a bridge to their authors. At the same time, however, in the process of their reproduction, they are transformed and translated into a modern form that combines historical characteristics with modern elements that can be called humanistic by the initiators of the reproductions. The interest in the visual form of the text in the manuscript also begins with Petrarch, as Otto Pächt has already pointed out 19 . His vivid criticism of contemporary scribal practice is often cited: He complained of «copyists who pride themselves on small, cramped lettering that baffles the eye; by heaping and cramming everything together, […] [their writing] confuses the spacing and piles up the letters, as though they were riding on the top of one another, so that the scribe himself could scarcely read them, were he to return a little later, while the patron who commissioned the book would really purchase not so much a book, as blindness because of the book» (Petrarch, 1992: 198).
Wayne H. Storey stresses the importance that layout had for the author in his analysis of the interaction between Petrarch's writing and the mise-en-page of his texts: For Petrarch not only was the design of the book a reflection of the edition's intellectual structure, it was also an integral part of its systems of meaning, from the clarity of its script to the unified organization of its knowledge in the text and its apparatus and glosses. It is, as Armando Petrucci has pointed out, Petrarch's preference for the simplicity and clarity 17 «Nochmal zeigt sich, dass der Verehrung eines Autors der abstrakte, immaterielle Charakter der Texte nicht genügt, genausowenig das allzu gegenständliche Buch. Gesucht ist ein körperliches Bild, dem der beim Lesen entstandene Affekt entgegengebracht werden kann » Löhr, 2011: 22. 18 Paris, Bibliothèque Nationale de France, ms. Lat. 8082, fol. 4r. 19 While Otto Pächt is not the only one who very early on raised the question of the aesthetic dimension of the emergence of the humanistic minuscule -these questions had already arisen in paleography in the 1920s -he is the first art historian to address this phenomenon of writing. See Pächt, 1957: 184-94. For an example of earlier approaches see Lehmann, 1918. of form that drove him to admire copies of the tenth and eleventh centuries in minuscule Caroline hands and to detest the often calligraphic and illegible minuscule forms of Gothic scripts overburdened by abbreviations and compendia, and tied to Scholastic thought (Storey, 2018: 17) 20 .
While legibility may be one important point in the development of the humanistic minuscule, the visible demarcation of this scholastic past must surely be considered as one major concern for Salutati, Poggio Bracciolini, and Niccolò Niccoli. They created a humanistic book written in the humanistic minuscule, littera antiqua, as they called the historic model. And the layout reflected the ideas of clarity and elegance first formulated by Petrarch (Storey, 2018: 17).

hamilton 166
The Latin text forms a regular bloc on the manuscript page. Only the added translation of a Greek word stands above the line. The parchment is clear and smooth. The text of Cicero's letter begins with a single decorated initial without further illustration or accentuation; a title page is absent. In the middle of this first page, there is a second letter: «Cicero bruto sal[utem]» (Fig. 7). A simple red capitalization without decoration offset from the rest of the text indicates where the letter begins. The separation of the text bloc from the frame in which the capital is inscribed is not only marked by the end of the text lines. Through embossing or debossing (Fig. 7), the capital is tangibly inscribed into the parchment by a double impression of the ruling. This rilievo gives the page a haptic dimension, for the viewer is invited to touch the surface to feel the trace of the ruling, a haptic dimension which functions in contrast to the smooth surface of the parchment itself. This three-dimensional demarcation of the text space accentuates the distance between the regular text lines written in dark ink and the surface of the parchment. On the following pages (Fig. 8), where wide white margins frame the even text blocks of the double pages on both sides, the effect of elegant restraint produced by the different compositional elements is even more accentuated. In the text blocks, it is the script that functions as a compositional element: the writing lines alternate with an interlinear space as large as the script itself. Moreover, the handwriting is upright without a trace of haste in the act of writing, which stands in contrast to the cursive script in the model used by Poggio (Fig. 9). Within this calm composition, the ascenders and descenders of the different letters introduce rhythmic cuts in the interlinear spaces by giving a vertical movement to the alternation of empty space and writing in the horizontal lines 21 .
The restrained decoration of these first pages continues. On 163 folios, the scribe only integrates 20 decorated initials. And even these are of modest character (Fig. 10): they are colored in red, blue, pale green, and yellow, showing the type of decoration called «bianchi girari» or «vinescroll decoration» typical of later humanist manuscripts 22 . These initials may have been based on a model found in a Carolingian manuscript from Salutati's collection: The manuscript Florence, Biblioteca Medicea Laurenziana, ms. San Marco 385 (Fig. 11) 23 could be viewed as their direct predecessor, as folio 1v contains an initial whose form and color are at first glance very similar to those found in Hamilton 166. But again, his interpretation is not mechanical. The hypothesis that these initials might have served as a source of inspiration for Poggio that he then turned into his own design is also supported by the fact that he did not imitate the forms and colors of any of the other initials from the same source. For instance, on folio 2v the initial «I» (Fig. 12) appears on a dark red background with small gold ornamentations each consisting of three points. A blue peacock with light-colored vine branches coming out of its beak also appears on the background. Neither this nor other figurative decorations can be found in the manuscript copies made by Poggio 24 . He also did not adopt the mise-en-page divided into two columns, instead opting for a uniform single column. The aesthetic seemed to have been so important to Poggio that he even added filling letters -in particu-21 For example, the vertical lines of the minuscule «d» or «p» and the elegant swing of the «g». 22 For the first art historical observations of the details of the decorated initials see Pächt, 1957: 189-93. More recent studies are: Ceccanti, 1996;Crivello, 2003;Mulas, 2014. 23 The rectangular cover page that can be seen in the digital version did not belong to the text's 12th-century layout. It was probably added in 1448 when the bookseller Vespasiano da Bisticci made other repairs to the manuscript. At this time, the book was already a part of the collection of the San Marco monastery library (see de la Mare, 1992: 188). 24 There are no references to this in the secondary literature either. However, not all manuscripts that fall within the purview of this work could be checked for its possible presence. lar the letters «o» and «I» -after proofreading the text, thus producing the uniformity of the text block at the expense of linguistic correctness (Ullman, 1974: 129).
In addition to the initials, Poggio uses a capitalis that shows some parallels to the display scripts used in Carolingian manuscripts (Bischoff, 1990:146), and very clear parallels to scripts used in ancient inscriptions 25 (Fig. 14). This script was used for the titles of books and was always used in combination with a decorated initial. The wide interlinear spaces before and after the titles mark a deviation from the otherwise uniform continuity of the regular text body and are the only element that does so in this copy.
The majority of the manuscript's pages feature uniform text, large unused frame space and simple initials. The letters as well as other elements on the manuscript page, such as ink lines on white parchment surfaces, thus become part of the manuscript's composition. Clarity, uniformity and restraint in the use of decoration define the visual appearance of this copy of Cicero's letters to Atticus, a collection rediscovered by Salutati, whose content and grammar was restored by Poggio.

Seeing Written Words
The concrete conceptual nature of Niccoli's and Poggio's interest in the material and visual dimensions of manuscripts becomes clearer when one considers a highly critical remark made by Guarino da Verona in 1412 (Davies, 1986: 61): Neglecting the other aspects of books as quite superfluous, he [Niccolò Niccoli] expends his interest and acumen on the points (or dots) in the manuscripts. As to the lines, how accurately, how copiously, how elegantly he discusses them. […] You would think you hear Diodorus or Ptolemy when he discusses with such precision that they should be drawn rather with an iron stylus than with a leaden one. […] As to the paper, that is the surface, his expertise is not to be dismissed and he displays his eloquence in praising or disapproving of it. What a vacuous way to spend so many years if the final fruit is a discussion of the shape of letters, the colour of paper and the varieties of ink […] 26 . 25 Poggio's extensive knowledge of this type of artifacts is evidenced by his study of antique inscriptions, which he documented in a sylloge compiled as early as 1404 during his first trip to Rome. However, as we only have partial copies of this sylloge from two 16th-century manuscripts, it is difficult to make any inferences about the visual interest shown in Poggio's studies in general, and if he studied the antique Capitalis in an imitative way in particular. 26 Cited from Gombrich, 1976: 97f.        The material aspects discussed by Guarino are those that appear on the very page of the manuscript. The critical note thus highlights the fact that Niccoli and Poggio reacted with particular acuity to the visual appearance of the materials they were working with.
The references to Diodorus and Ptolemy cited here by Guarino also suggest that the contemporary concern with the production of humanistic copies was not purely a matter of handicraft technique. Rather, Poggio's concept of a humanistic book is based on a multi-layered understanding of book, script, and mise-en-page. Coluccio Salutati, chancellor of Florence for 30 years, mentor of Poggio Bracciolini and an important figure among the early Florentine humanists, helped lay the groundwork for material interpretation through the library he made accessible to young humanists, as Poggio remembers in a letter to Niccoli after Salutati's death 27 , and for the theoretical interpretation of the visual appearance of manuscripts. Salutati apparently based his interest in manuscripts' appearances on a quote by Priscian, a grammarian of late antiquity. A manuscript containing his institutio de nomine, pronomine et verbo is the most ancient manuscript in Salutati's collection, a Carolingian example of the 9th century 28 . Salutati quotes him 29 : «The letter is as it were legitera, because it shows the path to readers» 30 .
At first glance this quote reveals a classical understanding of writing as a purely graphic trace of language, prominent throughout the Middle Ages and later. Before Salutati, it was cited by Petrarch, and the chancellor of Florence thus positions himself in the footsteps of his ancestors from antiquity 31 . This definition places an important accent on the legibility of texts, interpreting the letter as the visual trace of the spoken sound. At the same time, however, the path or journey -iter -which leads the reader through the book is mentioned. 32 As per Quintilian it is the ductus that the reader follows on this path through the text, and thus 27 «How can I fail to mention that he was a father shared by all and a friend of good men; all those in whom he perceived some gleam of intellect he not only fired with a zeal for virtue by his words but actually helped them far more with his resources and especially his own books, which he wished to be a cornucopia for other men's use as much as for his own», Letter II, cited from Gordan, 1974: 23. 28 Florence, Biblioteca Nazionale Centrale, Conventi Soppressi J.10.46. For a detailed description, see De Robertis, et al. 2008: 227-29, cat. no. 56. 29 «Years later, moreover, he confided that, around the same time, the study of Priscian's monumental text awakened him -and again he also credited divine influence -to the importance of orthography, initiating his lifelong concern with the reform of spelling», Witt, 2000: 295; with a reference to Ullman's research, see Ullman 1963: 108. 30 «Lettera est quasi legitera, quia legentibus iter praebet», cited from Amsler, 1989: 225. See also, Petrucci, 1995 See Witt, 2000, especially from p. 292 onwards. 32 «Ductus is the way by which a work leads someone through itself: that quality in a work's formal patterns which engages an audience and then sets a viewer or auditor or through the book. While the stylistic ductus leads one through the text, Quintilian describes the material counterpart to it, the litterarum ductus, which is the line that follows the writer's formalized handwriting (Carruthers, 2010: 195).
For him, both ducti lead one through the text, on the textual and the material level. This rhetorical function, which combines the reading of the script with the reading of the text, shows that the material potential of the written, i.e. the visual effect it produces before content is deciphered, can be used as a medium of expression. During the Middle Ages, this potential was increasingly exploited, especially in the context of Christian writing culture. Here, the differentiated representation of the text became unavoidable: the mise-en-page of differentiated texts of a different nature within the same book, often on the same page. Holy Scripture on the one hand, and its commentaries, ritual instructions, canonical tables, which clarified the references within Scripture in tabular form on the other, required a more complex use of typography, in addition to the use of different font sizes, fonts, and script colors. Complex systems using tables, glossaries, marginal texts and the like, in order to distinguish among the genres of text, came to be developed (Rouse & Rouse, 1982). This variation of forms and fonts gave the manuscript page a life of its own, which no longer only documented the text, but also depicted it in its relationship to other types of text. Both the basic idea that the Word of God became Scripture, thus giving great significance to the book as a vessel, and the development of varied ordering and illustrative possibilities for the mise-en-page, developed the auratic potential of the book and its pages (Martin & Vezin, 1990).
In addition to this symbolic dimension of the book as an object, the manner in which knowledge is ordered within the manuscript becomes increasingly structured 33 . The ordering of knowledge in the book was not exclusively textually documented in relation to earlier non-written orders of knowledge, as with mnemonics. Rather, knowledge architectures were created within the space of the codex. The book increasingly became a space of knowledge 34 . The knowledge inscribed in it was seen as an image of the macrocosm in the microcosm of the codex. Thus, the manuscript's visual design and its script were considered fundamental to performer in motion within its structures, an experience more like traveling through stages along a route than like perceiving a whole object», Carruthers, 2010: 190. 33 «Twelfth-century scholarship is characterized by the effort to gather, organize, and harmonize the legacy of the Christian past as it pertained to jurisprudence, theological doctrine, and Scripture», Rouse & Rouse, 1982: 201. 34 On the collecting and the new ordering of knowledge in the codex, see Meier, 2003: plates VIII-X; see also Meier, et al. 2002. the impact of the content on the reader, who was always simultaneously a beholder of its appearance.
Hugo St. Victor, for instance, drawing on the scholastic reading of the 12th century, writes in his Didascalicon that wisdom relates to religious illumination: «Sapientia illuminat hominem, ut seipsum agnoscat» 35 . For Hugo St. Victor wisdom inscribed into the parchment pages of the manuscript makes the material surface itself a source of wisdom. Simply looking at the pages of the manuscript illuminates the reader as if the pages were a mirror 36 .
Both medieval understandings of what a book is and how it can affect the reader, as well as various aspects from ancient reflections on writing and the presence of the author in his style, are united in the visual concepts of Poggio Bracciolini and Niccolò Niccoli. But could there be a logical formula through which the humanists could translate their literary and rhetorical interests into visual appearance? Michael Baxandall (1971) has very remarkably shown the interconnections between new artistic concepts and topoi and ancient literature on rhetoric. And we do not have to go too far to find a model for the humanistic aesthetic concept for manuscript layout, as we will see. In his late work, Orator, Cicero describes the elements of good style: In the range of the same style, some styles are very smart but unornate and deliberately adapted to the unpracticed, the unexperienced. Others with the same soberness seem more pleasing, that is complaisant, vivid and show flashes of very effective ornamentation. Halfway between these two stands the middle and well-balanced style. It does not have the exuberant presence of the last one, nor the flow of words of the first. This middle style neighbors both, not falling out of its frame on one side or the other, a part of both and better if we are searching for trueness, free of either. The oration flows, as one says, in a single stroke, presenting nothing other than sophistication and regularity. It adds a bow to the crest and enriches the whole speech with appropriate decorative expression or idea 37 . 35 «Wisdom illuminates man so that he may recognize himself», Taylor, 1961: 46. See also Illich, 1993 Although Salutati himself did not quote Hugo St. Victor's arguments, the latter's texts are contained in his collection, and the many annotations evidence an intense reading. Ullman quotes two manuscripts containing texts by Hugo St. Victor in his catalogue: Vatican City, Biblioteca Apostolica Vaticana, ms. Vat. Lat. 678 and Vatican City, Biblioteca Apostolica Vaticana, ms. Pal. Lat. 309. See Ullman, 1963: 180n73 and 194n99, respectively. 37 «In eodemque genere alii callidi, sed impoliti et consulto rudium similes et imperitorum, alii in eadem ieiunitate concinniores, id est faceti, florentes etiam et leviter ornati. Est autem quidam interiectus inter hos medius et quasi temperatus nec acumine posteriorum nec fulmine utens superiorum, vicinus amborum, in neutro excellens, A well-balanced style that does not fall out of its frame, oration in flow, characterized by regularity and appropriate decoration -the rhetorical characteristics of this well-balanced style are translated to the mise-en-page of the humanistic manuscript. And Poggio clearly points out his similar interest: The parchment which I ordered in folio size I want for transcribing the Verrine Orations in one volume and likewise in another volume the Tusculans and the De finibus bonorum et malorum; I want another set for the letters to atticus. Now reflect on it and see whether this measure will do for these volumes and make sure that it seems to suit to their elegance (Gordan, 1974: 93 [Letter XXXV]).

Conclusion
The addressee of the Hamilton 166 was the young Cosimo de' Medici, a friend of Salutati and Niccoli. His humanist interests made him a frequent guest of the humanist circles of Florence. At the same time, however, he remained a representative of his social and political position. The manuscript was not simply supposed to be an interesting work in Cosimo's collection. Rather, it served as a manifesto of Poggio and Niccoli's humanist ideas that was to be displayed in a prestigious and publicly significant collection. The manuscript's colophon makes it clear that Poggio had certainly considered the significance of this function of the copy (Fig. 15): He signed this copy for which he did much more than simply fulfill the function of the scribe. He forged a conceptual bridge between his humanist philological work and an aesthetic that through «puritas and suavitas» 38 made the purity of grammar that had been restored by the humanists visible in the very appearance of the script. The manuscripts that Poggio produced together with Niccoli deployed the visual appearance of the materials they worked with in order to create a visual manifesto: making works speak to the eyes 39 . utriusque particeps vel utriusque, si verum quaerimus, potius expers, isque uno tenore, ut aiunt, in dicendo fluit nihil afferens praeter facilitatem et aequabilitatem aut addit aliquos ut in corona toros omnemque orationem ornamentis modicis verborum sententiarumque distinguit», see Cicero, 2004: 31 (VI.20-22). A translation from the German version by the author is used here as the English translation (Cicero, 1962: 318-21) seems in contradiction with the more recent German one. 38 Traversari, 1968: Vol. 2, Lib. XI, 19. See also Meiss, 1960: 99;and Pfisterer, 2002: 94. 39 This formulation takes up the thesis of Volker Breidecker, who sees in the art of the city of Florence a «visual rhetoric». He thereby takes in the «visibile parlare» of Dante. See Breidecker, 1992: 9.