The text criticism or text-critical method (from ancient Greek κρίνω krínō “distinguish, sort out , select”) is a method with which the influences of the origin and transmission on the shape of traditional texts are worked out. It is used when there are different versions in order to (re) construct a canonical text version ( text edition ) from manuscripts or first prints. It belongs to edition philology , which in turn is a sub-area of literary studies . The purpose of an edition is usually a critical edition of the text that should be legible and at the same time note the changes and decisions made by the editors. Text criticism does not elaborate an interpretation of the text, but provides the material, which is then analyzed in terms of content in the exegesis or text interpretation .
The text-critical method
At the beginning of all text criticism is the search for the text witnesses , that is, manuscripts or early printed editions of the text. This task is much easier today than it was in the early days of printing, as most libraries and archives cataloged their manuscript holdings, documented a great deal on microfilms and photographs and recorded them as scans. The text critic no longer has to travel around the world and rummage through each library individually to find text witnesses and copy them individually by hand. The results of the text criticism depend largely on the selection, quality and completeness of the text witnesses.
The basis of text criticism is the collation , i.e. the comparison of several handwritten or printed versions of a text. The texts are sorted chronologically - if possible - and then compared word for word, sentence by sentence. In a first step, the archetype is reconstructed from this , i.e. the text version on which all preserved text witnesses are based. This is usually not identical to the author's original text, but can itself contain errors or changes to the original text, which are then tracked down and corrected in the second step if possible.
The first step takes place in four stages:
1. Heuristic : All text witnesses received are searched for and collected. Fragments and presumably earlier versions are recorded as well as secondary evidence, i.e. H. Quotations from the text in question from later authors. Translations are also secondary evidence, since depending on the translator's fidelity to the text, conclusions can be drawn from the translation, albeit indirectly, about the version available for the translation. The secondary witnesses are sometimes conceptually differentiated from the primary witnesses as “testimonies”.
3. Recensio : The variants are analyzed, particularly with regard to their emergence. If possible, a family tree ( stemma ) is created that provides information about which manuscript was copied from which. In the stemma, developed versions (not preserved intermediate stages, called hyparchetype ) are entered. Witnesses who are only copies of other surviving witnesses can be excluded from further consideration, "eliminated". If due to contamination (the transcription errors appear in the witnesses in a colorful mixture and do not reveal any family relationships), no family tree can be determined, one speaks of "open tradition"; in this case all text witnesses are initially of equal weight and eliminatio is not possible.
The following corruptions are typically found:
- Write-off oversight:
- duplicate sentences, lines, words or letters ( dittography ),
- omitted sentences, lines, words or letters ( haplography ), especially if sentences, words or phrases are repeated literally or almost literally,
- Confusion of letters with a similar shape or a similar pronunciation (such errors are often instructive for the dating of text versions, since different scripts and pronunciations have been in use over time),
- Spelling mistakes, spelling changes;
- a difficult text has been simplified,
- a short text was added.
- an uncommon text has been aligned with a common one.
4. Examinatio ("verification"): The quality of the readings is assessed according to the criteria of language, style and internal coherence; the result is the reconstruction of the archetype . Frequent argumentation patterns are:
- The more original reading is that which can best explain how the other readings came about. (This principle can be compared to the determination of the phylogeny of a living being in biology.)
- The principle of lectio difficilior applies , which means that the more difficult reading is probably the older. This principle is based on the assumption that when a text is copied it is simplified and smoothed out rather than rendered more complex and incomprehensible. Complicated sentences are simplified, outdated words, word forms and formulations are replaced by more modern ones. In the event of corruption, a text that is no longer understandable or obviously meaningless is changed to one that is easier to understand. At the same time, however, there is a risk of artificially complicating a text through the unreflective application of this principle. Proceeding according to lectio difficilior for texts that do not make high literary demands on themselves becomes completely problematic, if not pointless .
- The older a text witness, the fewer transcription errors it is likely to contain. It must be taken into account that more recent copies may have had very old or high-quality originals and that, conversely, very old manuscripts can also be of inferior quality if they were made by an incompetent scribe.
Where the text version determined in this way is incorrect, the second step is to try to restore the original text of the author through divination (wise and well-founded guess) ( emendation ). Here, too, the editor judges based on his knowledge of the historical context, the language used by the author and his time, the internal structure of the text to be edited, and its embedding in the literary environment (intertextual references). Two means are available:
- Conjecture : The editor replaces the text of the archetype with another one that has nowhere been handed down. Often, but not always, the conjecture in the typeface resembles the traditional text, so that a plausible hypothesis can easily be given regarding a possible prescription process. Conjectures can have completely different degrees of certainty, from the highly speculative suggestion, which is more used to draw attention to a problematic point ("diagnostic conjecture"), to the obvious and beyond doubt correction, a so-called coniectura palmaris (" obvious conjecture ", from palma ," palm ").
- Athetesis : the editor removes text that he has recognized as a later ingredient and which therefore did not belong to the original text of the author; this is indicated by square brackets. Often these are marginal notes, explanations, or comments by a scribe that a later copyist saw as part of the text.
Conversely, the editor can also be of the opinion that the original text has failed ( Lacuna ); Handwriting sometimes has holes in the writing material or parts of the page are destroyed or torn off, individual letters or words are overwritten, blurred or no longer legible. Completed words or letters are put in angle brackets, longer gaps are usually indicated with asterisks. Only in rare cases are longer gaps filled with appropriately invented text. In the case of larger gaps, the editor can give a summary of what was presumably contained in the lost text.
The result of the text-critical method is an educated guess as to which text the author might have written. The heart of a scientific edition is the reconstructed text, but of decisive importance is the text-critical apparatus - mostly at the foot of the text - which documents the divergences between the individual text witnesses. Thus the reader can on the one hand follow the actions of the editor, on the other hand make their own considerations about the reconstruction of the text, if z. B. different readings result in different directions of meaning. Ultimately, there can be no undisputed reconstruction of a text if there is more than one tradition.
There are two possible forms of a text-critical apparatus:
- Negative apparatus: The apparatus only records the readings of the text witnesses that deviate from the constituted text.
- Positive apparatus: All readings - including those chosen for the constitution of the text - are written out here.
Characters that can appear in the text:
- †… † : The text is so badly spoiled that it can no longer be deciphered, perhaps because the scribe wrote too indistinctly, a scraping was carried out or the material is damaged. The three points show that the publisher does not want to make any statements or assumptions about what could have been here (so-called crux critica or crux desperationis ).
- † Text † : One and the same text has been handed down in all text witnesses, but it does not make any syntactic or semantic sense, and none of the previously suggested conjectures has sufficient probability in the editor's opinion.
- 〈Text〉 : The editor herewith supplements text that is not attested in any textual witness. The suggestion does not have to come from him; it can also have been made by another philologist .
- [Text] : The editor considers the text bracketed in this way, which is attested in many or all text witnesses, to be not original.
Text criticism of different types of text
The basics of text criticism apply to all types of text. For different types of texts, however, there are different problems, some of which require different methods or focuses.
Texts from classical antiquity
There are often only a few text witnesses for the publication of texts from antiquity, and they are many centuries younger than the original text. The best attested work is Homer's Iliad with 700 text witnesses. The collation is a manageable task here. On the other hand, the emendation plays an important role, as one often cannot assume that the original form is actually contained in one of the few variants.
For the Hebrew Bible , the Masoretic text is a text type preserved in many medieval manuscripts and passed down very precisely with the help of the Masora , the existence of which has been secured for the 1st century by "protomasoretic" text fragments found in Massada , Wadi Murabbaʿat and Naḥal Ḥever is. Other Hebrew text witnesses are the Samaritan Pentateuch and the manuscripts of biblical books found in Qumran , which, apart from the Great Isaiah Scroll, have only survived in fragments.
While, in the opinion of most text critics, the Masoretic text type comes very close to the original text in the Torah , this cannot be assumed in other books, such as the Samuel or the Jeremiah . In view of the sparse non-Masoretic Hebrew text tradition, the ancient translations, especially the Greek Septuagint and its daughter translations, e.g. B. the Vetus Latina , of immense importance for textual criticism. There is no doubt that many of the variants only attested in these translations go back to Hebrew models, because some Qumran manuscripts , such as 4QSam a or 4QJer b , testify to Hebrew readings that deviate from the Masoretic text, but at the same time from the Hebrew Septuagint model that was only suspected until then correspond.
Naturally, there are great uncertainties when the text is corrected in the original language with the help of a translation. That is why most of the critical editions of the Hebrew Old Testament, such as the Biblia Hebraica Stuttgartensia or the Biblia Hebraica Quinta , diplomatically print the Masoretic text and merely refer to the variants in the apparatus .
The New Testament is a special case for textual criticism due to the much higher number of text witnesses. There are over 5,000 Greek text witnesses, over 10,000 Latin manuscripts and a further 10,000 manuscripts from translations into other languages, plus countless quotations in other scripts. Several templates were often used to create New Testament manuscripts, so that one manuscript can have several mothers. This practice makes stemming very difficult and sometimes impossible. New Testament text critics take this special situation into account by grouping the text witnesses into text types based on recurring similarities in the text version and proceeding eclectically when choosing the variants. The main text types are the Alexandrian , Western, and Byzantine text types .
Due to the large number of early text witnesses, it can be assumed for the New Testament - in contrast to other ancient texts - that the original text form has been preserved for each individual passage in at least one manuscript. Conjectures therefore no longer play a role in New Testament textual criticism.
Texts from modern times
Textual criticism is also used in modern literature when different versions of a text exist. Special questions arise when different manuscripts from the author's lifetime contain different variants.
History of Textual Criticism
Even in antiquity there were editions of texts with which one tried to come as close as possible to the original text. The Alexandria library, for example, is considered the production site for many Greek classics. Most of the time, modern textual criticism is only able to determine this text form, which was standardized in antiquity, because it is the last (often only fragmentary) version of a text that has been preserved.
In the Middle Ages , especially in the Byzantine Empire, the old classics were cultivated and bad variants were singled out. At this time, Judaism also tried very hard to pass on its holy scriptures unadulterated, which culminated in the Masoretes between 780 and 930.
Today's text-critical method was developed in the 19th century by classical philology in order to reconstruct ancient texts (some of which are only fragmentary or in very late copies, but have been handed down in several lines of tradition). The philologists Friedrich August Wolf , Karl Lachmann and FDE Schleiermacher made outstanding contributions to their methodology .
Criticism of textual criticism
Joseph Bédier criticized the stemmatic method as early as 1928 when he examined medieval French manuscripts that had been classified according to this method. He noted that most of the traditions were divided into two branches, although there was, in fact, no reason why three- or four-branch traditions should be less common. He concluded that this method did not meet strict scientific standards, could not correctly reflect the actual history of the text and contained too much subjective leeway. Paul Maas , on the other hand, defended the stemmatic method as early as 1937, pointing out that a three-column stemma is actually much less likely than a two-column one, since among the 22 different theoretically conceivable types of the stemmatic relationship of three related text witnesses, only one is three-columned.
Bernard Cerquiglini emphasized in recent years that the tradition of the vernacular medieval literatures (Old French, Middle English, Middle High German) cannot be compared with that of the Latin and Greek "classics" and the sacred texts and therefore the method of textual criticism cannot be applied to them be. Medieval literature is a literature of variants, in which an “ original text ” or the exact reproduction of a template played hardly a role. The aim of creating an original text applied modern notions of copyright and authorship to old texts without understanding the medieval background.
Some recent text-theoretical and editorial views question the primacy of the search for the original text as a whole. Text criticism is used here as a means of analyzing the tradition, but the construction of a text beyond the actually existing documents is rejected as a dehistoricization. In these schools, the authenticity of the tradition takes precedence over the arbitrarily claimed authority of an editorial text setting. For the Anglo-Saxon editor David Greetham summarizes the rejection of a critically constituted text and, above all, the mixture of witnesses to an eclectic text in the catchphrase of "text that never was".
General literary studies
- David C. Greetham: Editorial and Critical Theory. From Modernism to Postmodernism. In: George Bornstein, Ralph G. Williams (Eds.): Palimpsest. Editorial Theory in the Humanities. University of Michigan Press, Ann Arbor MI 1993, ISBN 0-472-10371-7 , pp. 9-28.
- Herbert Kraft : Edition Philology. 2nd, revised and expanded edition. Lang, Frankfurt am Main et al. 2001, ISBN 3-631-35676-5 .
- Jerome J. McGann: A critique of modern textual criticism. University of Chicago Press, Chicago IL 1983, ISBN 0-226-55851-7 .
- Werner Schröder: Smaller fonts. Volume 6: Text transmission and text criticism. 1965-1993. Hirzel, Stuttgart 1994, ISBN 3-7776-0587-5 .
- Georg Witkowski : Text criticism and editing technique of more recent written works. A methodological attempt. Haessel , Leipzig 1924.
- Thomas Bein: Textual Criticism. An introduction to the basics of Germanistic-Medieval edition studies. Textbook with an exercise section. Peter Lang, Frankfurt am Main et al. 2008, ISBN 978-3-631-56160-7 .
- Klaus Grubmüller: Edition. In: Reallexikon der Germanischen Altertumskunde . Volume 6: Donar - Dugout. 2nd completely revised and greatly expanded edition. de Gruyter, Berlin et al. 1986, ISBN 3-11-010468-7 , pp. 447-452.
- Josef Delz : Text criticism and editing technique. In: Fritz Graf (ed.): Introduction to Latin Philology. Teubner, Stuttgart et al. 1997, ISBN 3-519-07434-6 , pp. 51-73.
- Edward J. Kenney : The Classical Text. Aspects of Editing in the Age of the Printed Book . University of California Press, Berkeley-Los Angeles-London 1974, ISBN 0-520-02711-6 .
- Paul Maas : Textual Criticism. 4th edition. Teubner, Leipzig 1960.
- Giorgio Pasquali : Storia della tradizione e critica del testo. Le Lettere, Florence 1988, first edition 1934.
- Pöhlmann, Egert: Introduction to the history of transmission and the textual criticism of ancient literature ; Scientific Book Society, Darmstadt 1994–2003; 2 volumes, volume 1: Antiquity ; Volume 2: Middle Ages and Modern Times , by Egert Pöhlmann , with contributions by Christianwirt, Paul Klopsch and Georg Heldmann; ( Ancient Studies ); ISBN 3-534-04495-9 , 3-534-12440-5.
Sebastiano Timpanaro : La genesi del metodo del Lachmann. Le Monnier, Florence, 1963; nuova edizione riveduta e ampliata, Liviana, Padua 1981; most recently: UTET, Turin 2004.
- German translation: The origin of the Lachmann method. Second, expanded and revised edition. Authorized translation from Italian by Dieter Irmer . Buske, Hamburg 1971. (For the German edition extended and revised by the author)
- English translation: The genesis of Lachmann's method. Edited and translated by Glenn W. Most . University of Chicago Press, Chicago 2005. Publisher Ad , Google Book Preview
- Martin L. West : Textual Criticism and Editorial Technique. Applicable to Greek and Latin Texts (= Teubner Study Books. Philology. ). Teubner, Stuttgart 1973, ISBN 3-519-07401-X .
- Siegfried Kreuzer : Text Criticism. In: Siegfried Kreuzer, Dieter Vieweger and others: Proseminar I. Old Testament. A work book. 2nd, revised and expanded edition. Kohlhammer, Stuttgart et al. 2005, ISBN 3-17-019063-6 , pp. 26-48.
- Emanuel Tov : Textual Criticism of the Hebrew Bible. 3rd, revised and expanded edition. Minneapolis, Fortress Press, 2012, ISBN 978-0-8006-9664-1 .
- Editio. International yearbook for edition science. Vol. 1, 1987 ff., .
- Text. critical contributions. Vol. 1, 1995 ff., .
- The European Society for Textual Scholarship.
- An Online Textual Commentary on the Greek Gospels Application of text-critical principles to the text of the Gospels (English, several PDF files with a total of around 2000 pages)
- Emanuel Tov (Hebrew University, Jerusalem): Electronic Resources Relevant to the Textual Criticism of Hebrew Scripture (English)
- Alexander Achilles Fischer: Bible text / text criticism . In: Michaela Bauks, Klaus Koenen, Stefan Alkier (Eds.): The Scientific Biblical Lexicon on the Internet (WiBiLex), Stuttgart 2006 ff.
- Peter Riemer , Michael Weißenberger , Bernhard Zimmermann : Introduction to the study of Latin studies. 2nd, completely revised edition. Beck, Munich 2008, ISBN 978-3-406-44051-9 , p. 69.
- Paul Maas, Leifehler and stemmatic types (1937), in: Ders., Textkritik, 4th edition, Leipzig 1960, pp. 26–32, here p. 29. Of the other 21 options, 15 are two-columned and 6 are single-columned.