<oo>→<med> Medieval Rus′

Maintained by: David J. Birnbaum (djbpitt@gmail.com) Last modified: 2020-08-30T19:23:33+0000

Textual criticism: terms methods, and principles

Ways of talking about textual criticism

• Textual criticism, text criticism Used here to refer to stemmatic textual criticism
• Textology May be synonymous with textual criticism, but not to be confused with Lixačev’s tekstologija
• Source criticism Identification of citations, paraphrase and other forms of text reuse; also identification of individual texts within a compilation.

Assumptions underlying stemmatic textual criticism

1. Not all witnesses have equal value as evidence for an earlier reading. Witnesses must be evaluated; we cannot just take a vote. (Manuscripts should be weighed, not counted. Kurt Aland and Barbara Aland, The text of the New Testament, 275–76.) The evaluation must be based on principles, rather than on intuition.
2. Where all witnesses agree, the reading is true (subject to amendment by conjecture).
3. Where witnesses diverge, exactly one reading is true (subject to amendment by conjecture).
4. A true reading is transmitted directly, unless there is evidence that it represents a correction of an error.
5. Shared errors have a common source (Maas: peculiar error = error that arises at a particular locus in the tradition). (If two people are found shot dead in the same house at the same time, it is indeed possible that they have been shot by different persons for different reasons, but it would be foolish to make that our initial assumption. James Willis, Latin textual criticism, Chicago, 1972, 14, qtd. in Ostrowski 1981, 15.)
6. Either one true reading can be identified or the evidence presents a crux.
7. The original may have contained errors.
8. Not all types of texts are amenable to stemmatic analysis, and where stemmatic analysis is appropriate, the stemma for one part of a text may differ from the stemma for a different part.

Principal PVL witnesses

From the Introduction to the HURI edition of the PVL, xx

• Laurentian (Lav, dated 1377)
• Academy (Aka, dated to the end of the fifteenth century)
• Radziwill (Rad, datable to the 1490s)
• Hypatian (Ipa, dated to ca. 1425)
• Xlebnikov (Xle, dated to the sixteenth century)

What to do (Maas)

• Recensio Collation and comparison of manuscript evidence and their grouping into families (recensions) to explore the transmission
  • Fontes criticae Identify the manuscript witnesses that bear independent witness
  • Eliminatio codicum descriptorum Elimate manuscripts that do not bear independent witness
  • Collatio codicum Collate the witnesses to align and compare the moments of variation
  • Examinatio Examine and evaluate the moments of variation (loci critici)
  • Selectio Choose among variants. Common errors are attributed to a hyparchetype
  • Constitutio stemmatis codicum Create a genealogical model of transmission (stemma codicum). Stemmatic textual criticism is associated with Karl Lachmann and has been developed extensively by others; the traditional handbook is Maas.
• Constitutio textus Come as close as possible to restoring the original. Only reaches youngest common ancestor.
  • Examinatio et selectio were used during recensio to arrive at a stemma; here variation is examined again to determine a best reading.
  • Restitutio Reconstruction (or, perhaps more accurately, construction) of a text by ascertaining the best reading where possible.
  • Emendatio ope codicum Choosing a true reading under the guidance of the stemma.
  • Emendatio ope ingenii Emendation that goes beyond the selection of a variant by stemmatic principles. Also called divinatio or conjecture.
  • Paradosis Best reading, as determined by textual criticism.

Textology and editions

• Diplomatic edition An orthographically precise transcription of a single manuscript source.
• Critical edition, text-critical edition An edition the prioritizes comparing variant readings across witnesses.
• Copy text, control texts In a traditional critical edition, the copy text (sometimes called base text) is the full transcribed text of the witness that the editor considers the best, that is, the closest to our understanding of the original. Variant readings are recorded in footnotes; the footnotes as a whole are called a critical apparatus (apparatus criticus); the manuscripts from which the variants are cited are called control texts.
• Dynamic criticial edition Instead of selecting one witness as the copy text, the editor chooses readings from different witnesses according to specific editorial principles and constructs a composite, eclectic text that serves as the copy text for the edition.
• Interlinear collation An interlinear collation (Umschrift-Partitur edition) presents full parallel readings from all witnesses, aligned for ease of comparison. The 2003 HURI edition of the PVL is an interlinear collation.
• How can a digital edition be more than a paper edition read on line?

Other types or properties of editions include: facsimile (images, rather than characters; scanned, photographic, lithographic, etc.), modernized (updates archaic features, usually by standardizing spelling), modern (translation from archaic to modern language), educational (for students, rather than scholars; often the copy text from a critical edition without the apparatus, typically heavily annotated for pedagogical purposes), interpretative (rigorously commented; often used for a codex unicus), etc.

Terminology

• Work Abstract term for a composition, either wholly original or wholly or partially based on other compositions, that can be identified by the presence of certain textual material and by the deliberate arrangement of this material according to a specific authorial program. Different works may share common textual material. The term is necessarily subjective, since often there is not an unambiguous point at which all scholars would agree that differences are sufficient for two texts to be considered to represent different works.
• Text Abstract term for each representative of a work, embodied in a manuscript. A manuscript may contain more than one text.
• Shared textual material Useful way of discussing textual transmission below the level of identifiable integral texts and works
• Recension Family, or group of witnesses related through shared innovations. Orthographic, paleographic, or linguistic term (e.g., Russian recension), comparable in meaning to redaction (as used in these disciplines).
• Family Group of witnesses related by shared innovations. Also called recension or branch.
• Redaction Abstract term that provides a means of subdividing the representatives of a work according to important changes in content. One may properly speak of two redactions of a single work if they contain substantially the same material in substantially the same arrangement and represent substantially the same authorial program. If the material, arrangement, or program is fundamentally different, it may be more useful to speak of two separate works that share common textual material. By convention, the original authorial text is the first redaction, although numerical designations are confusing because they don’t distinguish siblings from children. Orthographic modifications along the lines of what would be considered copy editing today represent an attempt to write correctly, rather than to alter the content of a work, and do not create a new redaction; this extends to lexical modification and even minor textual emendation (e.g., Solomon sat on his father[ David]’s throne).
• Variant Abstract term for subdividing the representatives of a work or redaction on the basis of systematic minor lexical, linguistic, or orthographic features. Sometimes called izvod or, among paleographers, recension (as in Russian recension, Bulgarian recension, etc.). In this sense, the term has little textological significance by itself. Variants may belong to a single linguistic or cultural community. The term variant also refers to individual differences at particular locations in a text, which is a textological concept. Substantive (significant) variants are important for a critical edition, while accidental (insignificant) variants are the sort that might have arisen by accident, and that do not provide evidence for textual transmission. For example, the presence or absence of a word would usually be considered significant; whether or not a word is abbreviated usually would not be considered significant.
• Witness Manuscript, but also used to designate the text represented by the manuscript. Likewise copy.
• Authority Independent vs. dependent authority refers to whether a manuscript may transmit true readings without being fully dependent on another extant witness. Witnesses without independent authority are eliminated from the collation (eliminatio codicum descriptorum) because they cannot provide evidence for the archetype that is not already provided by an ancestor.
• Protograph, antigraph, apograph Protograph (holograph, autograph) = author’s original text. The manuscript from which something was copied is the antigraph of the copy. A copy made from another manuscript is an apograph of that antigraph.
• Archetype, hyparchetype The archetype is the most distant common ancestor that can be retrieved by text-critical methods (which is not necessarily the authorial text). Hyparchetype is an intermediate common ancestor of part of the tradition. Some use archetype to mean author’s original.
• Exemplar The text from which scribe a makes his copy. May or may not be extant.
• Open tradition (horizontal transmission) and closed tradition (vertical transmission) Vertical transmission is from parent to child, with no contamination. In an open tradition, textual witnesses may draw on multiple sources. A stemma works best with a closed tradition; more than a small amount of contamination may preclude the effective construction of a stemma.
• Contamination, confluence Use by a scribe of more than one source, resulting in horizontal transmission. Also called contaminatio. Contamination may result from copying marginalia introduced into a copy by a reader, or by emendation from memory (especially in the case of well-known texts), rather than from drawing directly from multiple sources.
• Error Deviation from the original, even if the original is erroneous and the deviation correct in a more traditional sense of the terms. Hermann Kantorowicz: true = uninterrupted descent, while correct may be restored.
• Crux A situation where a best reading cannot be determined by stemmatic analysis, requiring recourse to other evaluative methods (see Principles, below).
• Codex unicus Unique witness. In such situations one typically transcribes the text as it appears, sometimes amended by conjecture. There is no evaluation of variation because, in the absence of more than one witness, there is no variation.
• Lectio singularis (pl lectiones singulares) Unique readings. Not useful for text criticism unless they represent an entire branch; hence eliminatio lectionum singularium.
• Greek and Latin letters In stemmata and critical apparatus, often lower-case Greek sigla (sg. siglum) represent hypothetical texts, upper-case Latin letters represent extant manuscripts, and lower-case Latin letters represent lost texts that are known to have existed at one time.
• Paradosis Best reading, in the sense of most likely to have been the ancestor of all attested readings. Also called alpha. This may be later than the authorial text, and editors may be able to impute readings to an authorial text on the basis of external evidence beyond what the direct manuscript evidence allows us to reconstruct for the paradosis.
• Basic Text Lixačev’s osnovnoj tekst. See his Tekstologija 495–96 and Ostrowski’s objections 1981: 25–26.

Principles

• Recentiores, non deteriores Younger manuscripts are not automatically textually secondary or otherwise inferior. A younger manuscript may witness an older stage of the text than an older witness. This principle is associated with Giorgio Pasquali.
• Lectio difficilior probior Prefer the more difficult reading.
• Lectio brevior potior Favor the briefer reading.
• Usus scribendi Editors should be sensitive to norms and habits that are typical of times or places or schools of writing.
• Choose the reading that explains the others Copying errors are usually mechanical but can also result from scribe’s misunderstanding the exemplar. When those possibilities have been eliminated or are unlikely, then one must consider the likelihood that variant readings are the result of scribes independently trying to correct what they perceive to be an error in their common exemplar. (Ostrowski 2009: 187)

References

• Lixačev Lixačev, Dmitrij Sergeevič. Tekstologija: na materiale russkoj literatury X–XVII vekov. 2nd ed. Leningrad: Nauka, 1983.
• Maas Paul Maas. Textual criticism. Tr. Barbara Flower. Oxford: Clarendon. 1958. (Originally published as Textkritik in 1927 as Part VII of Gercke-Norden, Einleitung in die Altertumswissenschaft, Vol. I, 3rd ed. Second edition Leipzig: B. G. Teubner, 1949. Third edition Leipzig: B. G. Teubner, 1957.)
• Ostrowski 1981 Donald Ostrowski. Textual Criticism and the Povest′ vremennykh let: Some Theoretical Considerations. Harvard Ukrainian studies V: 1. March 1981. 11–31.
• Ostrowski 2009 Donald Ostrowski. The application of biblical exegesis to the study of Rus′ chronicles. Medieval Slavonic studies (New perspectives for research) - Études slaves médiévales (nouvelles perspectives de recherche). Ed. Juan Antonio Alvarez-Pedroza and Susana Torres Prieto. Collection historique de l'Institut d'études slaves, 43. Paris: Institut d’Études Slaves, 2009. 169–91.
• Pasquali Giorgio Pasquali. Storia della tradizione e critica del testo. Florence: Felice le Monnier, 1934.
• Trovato, Paolo. Everything you always wanted to know about Lachmann’s method. A non-standard handbook of genealogical textual criticism in the age of post-structuralism, cladistics, and copy-text. Tr. Federico Poole. Revised ed. Padua: Webster srl, 2017.

I am grateful to Helena Bermúdez Sabel for comments and suggestions.