Transcription errors and typologies: A contribution

ABSTRACT

This article reflects on the concept of transcription errors in Textual Criticism and makes a critical review of the main typologies of errors. It adopts a theoretical perspective, and it proposes a functional typology of transcription errors that will serve, in the future, to provide general considerations about the transcription process and allow the description of the process of textual transmission within the scope of a theory of writing.

KEYWORDS:

• Textual Criticism
• error
• textual transmission
• terminology
• typology

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Notes

¹ For a critical review of the history and theory of the method, see Trovato 2014.

² Cunha (1985: 418) presented a classification of six different types of these errors (errors caused by distraction; errors caused by memory defects; errors caused by cultural limits; translation errors; and citation errors), which can be further reduced to the two types pointed out above.

³ Note that the author can make transcription errors whenever he transfers the text of a witness to a new writing surface. In this case he produces an autograph copy and, as a copyist of himself, he is subject to the same phenomena as other copyists. See, in this regard Pereira 2018.

⁴ This is especially easy if the variation takes place in a context of low alternation. I consider a context of low alternation to be one in which only two or three variants are admitted, without inconsistency and without a change in meaning. For example, in the phrase The feline that slept at the window had a yellow stripe along its tail, the colour of the stripe alternates from 1 to 5, as it can vary between all the colours that can be found on a cat’s tail (white, brown, black, grey, yellow). However feline has a low alternation, since it can only be replaced by cat.

⁵ I clearly distinguish between good reading (one that can be recognised immediately as grammatically, linguistically, and logically consistent, as opposed to evident error) and genuine reading (one that can be attributed to the autograph and which can be erroneous, since the autograph can contain errors, due either to the copyist or to the author). On this distinction, see Sobral 2016.

⁶ In the history of Textual Criticism applied to textual transmission, some authors equate error with innovation (Roncaglia 1975; Avalle 2002). In fact, any operation practiced on the text that leads to its modification results in an innovation. Conceptually, therefore, transcription errors are innovations. On the other hand, the transmission of texts (with an emphasis on medieval texts in vernacular languages) is not always motivated by the objective of reproduction, but rather, and frequently, by that of rewriting (which generates a new version of the same text), within the scope of which operations that lead to its modification are preformed judiciously, resulting in innovations. Although conceptually transcription errors are innovations, the term has been used more narrowly to designate rewriting operations that result in different versions of a text.

⁷ It was not possible to find the work in the Bibliothèque Nationale de France catalogue or in any other catalogue or record available. For information on Desrousseaux, see Dain 1956: 4.

⁸ ‘Dans le domaine grec, le maître incontesté fut A.-M. Desrousseaux. (…) C’est a lui que j’emprunterai l’analyse psychologique de l’acte de copie’ (Dain 1975: 41).

⁹ Dain uses the term opérations, and rejects étapes, ‘car tout se fait en même temps … ’, when we consider the copy as an act as a whole (Dain 1975: 41). I consider transcription not as an act but as a process, which includes several acts (I prefer this term to operations).

¹⁰ ‘Tel est le mécanisme psycologique du travail du copiste. Telle est l’explication à donner aux fautes de copie. Ces fautes, Louis Havet les a dûment classées et étudiées: fautes visuelles, influence du modèle, du contexte, de la personalité du copiste, etc. Je n’ai pas à reprendre cet exposé magistral, désormais classique’ (Dain 1975: 46).

¹¹ ‘ … changes made by scribes in copying. It must be emphasized that many of these are not visual but phonetic or psychological in origin’ (West 1973: 20).

¹² Even in 2014, Trovato reproduced what seems to be Dain’s error coefficient: ‘For simplicity’s sake … let us assume an average rate of one error per page’ (52).

¹³ There is no reason for this attribution of the modifying categories to Aristotle. They are in fact from Quintilian’s Institutio Oratoria (‘mistakes in writing are caused by addition or omission, substitution or transposition’, book I, The Institutio Oratoria 1989: 81), from which all the authors (especially the humanists) repeat them. In fact, Aristotle did not write a word about these categories.

¹⁴ Secondary elements of the discourse are grammatical operators (prepositions, conjunctions, articles, determinants), elements of enumeration including redundant duplications or triplications, and redundant elements in general. The lapse of secondary elements occurs frequently in the oral reproduction of statements, in a context in which there is neither reading nor writing, so we can attribute them to memorisation. In this regard, see the examples analysed in Timpanaro’s classic book, Il lapsus freudiano: psicanalisi e critica testuale (Timpanaro 2002).

¹⁵ Error consisting of the omission of a segment of letters that is repeated within a word or a short sequence of words. For example, E uyho star instead of E uyho o star; desuayradas instead of de desuayradas; se cõtem este falamẽto instead of se cõtem em este falamẽto (Orto do Esposo, Ms. Alc. 198, National Library of Lisbon, fls. 11a, 23a, 23d). Orto do Esposo is a fourteenth-century moral and didactic work in Portuguese that was transmitted by two manuscripts now held in the National Library of Portugal (mss. Alc. 198 and Alc. 212). Both manuscripts were copied at the Monastery of Santa Maria de Alcobaça (Portugal).

¹⁶ Error consisting of the duplication of a letter or group of letters within a word or a short section of words. For example: castradro instead of castrado; ẽtentẽder instead of ẽtender (Alc. 212, fls. 23a, 33a).

¹⁷ Error that consists of replacing one element of a word with another element of the same or contiguous word. For example: testemẽte instead of tostemẽte; popreza instead of pobreza (Alc. 198, fls. 19a, 24a).

¹⁸ I will use the following symbols to represent the copyist’s amendments: <xxx> (segment cancelled), <xxx>/yyy\ (segment amended by overlap), [↑ xxx] (segment added on the overline).

¹⁹ National Library of Portugal, Ms. Il. 219, fl. 15b. The manuscript was produced in the feminine Monastery of Jesus, in Aveiro.

²⁰ Only codices unicus are an inappropriate object for this analysis, since only obvious errors can be identified in them, thus limiting the validity of the conclusions.

Additional information

Notes on contributors

Cristina Sobral

Cristina Sobral is a researcher in Textual Criticism at the University of Lisbon. She is the author of several critical editions of ancient and modern literary texts and published articles about transmission of medieval Portuguese works. She also coordinates the Project Corpus of Ancient Texts in Portuguese up to 1520 (http://teitok.clul.ul.pt/cta/) and she is a member of Philobiblon/Bitagap team (https://bancroft.berkeley.edu/philobiblon/bitagap_po.html).