Error Tagging Taxonomy

Created by Věra Fialová (Masters Dissertation 2012)

Short link to this page: http://bit.ly/error_tag_VF

One

• Spelling – violations of conventions for representing phonemes by means of graphemes

• Capitalization: omission

• Capitalization: addition

• Punctuation: omission

• Punctuation: addition

• Punctuation: misselection

Two

• Case – includes problems with the genitive including the uses of the ‘noun + noun’ or ‘noun of noun’ combinations instead of the genitive, and confusion between the subject and objects of pronouns)

• Gender – errors involving grammatical gender in personal, possessive and relative pronouns)

• Number – errors caused by the addition or omission of the plural morpheme or its misformation)

• Person – primarily the omission of the third-person -s in the simple present tense)

• Comparison – errors involved in the use of the comparative and superlative forms of adjective and adverbs

• Negation – any error involving negation including the wrong use of negative adverbs such as rarely or hardly, or negatives involving no/any and their compounds (nobody, anything etc.)

• Tense/Aspect – misselection as well as misformation; includes forms expressing futurity

• Voice – problems connected with the passive versus active construction

• Word Class – errors concerning wrong word class, i.e. an adjective instead of an adverb

• Phrase structure – addition, omission, misselection or misformation of a component of any phrase within a sentence, excluding the errors covered by the grammatical categories listed so far

• Sentence structure – errors relating to the structural elements of a sentence and their order; this category does not include complementation patterns (covered with the lexico-grammatical domain) as well as the errors in sequencing of information in a sentence that affect cohesion (covered within the textual domain)

Three

• Word formation – errors that appear to be caused by wrong word-formation processes, above all affixation and compounding

• Countability – errors related to the countable versus uncountable property of nouns

• Complementation: further subcategorized within this level of annotation based on part of speech of the element that triggers the wrong complementation pattern

• Dependent prepositions: further subcategorized within this level of annotation based on part of speech of the element that triggers the use of the wrong preposition

• Unnatural chunk – unidiomatic-sounding sequence of words where the error is not easily pinpointed; it does not involve any error that could be categorized within the grammatical or lexical domain

Four

• Word form – misformations that result in non-existent words, excluding wrong affixation and spelling mistakes; covers also complex word forms

• Lexical phrase: form – misselection, addition or omission of an element within an identifiable (semi-)fixed phrase; in the next level of annotation, part of speech is assigned to the wrong or missing element

• Denotation – errors related to the conceptual meaning of words and phrases; unit identification then distinguishes between words, to which parts of speech are assigned, and phrases

• Collocation – errors connected with the collocational properties of words; concerns weak to strong collocations; part of speech is assigned to the word that has been misselected if distinguishable

• Connotation – errors related to the extended/secondary meanings of words, which might represent associations shared within a particular culture (this category thus borders with the sociolinguistic competence)

Five

• Cohesion: further subcategorized within this level of annotation based on the nature of the cohesive devise involved in the error (including the sequencing of information in a sentence); cohesion errors manifest themselves by wrongly marked semantic relationships between pieces of text or by discontinuity of the text despite the underlying relatedness of its propositions; the unit identification is based on the wrongly used or omitted cohesive marker

• Logical structure – involves pieces of text that are not logically organized in terms of the underlying conceptual relations between sentences or paragraphs; on the level of unit identification, the unit that needs to be restructured/would be affected by the recommended restructuring is selected

• Formal structure – concerns errors related to the conventional structure and formal features required by the specific text type or the written mode of discourse in general, usually in terms of the layout of the text

• Relational coherence – covers errors of conceptual unrelatedness of propositions within a text; cannot be fixed by restructuring

• Topical coherence – targets digressions from the main topic or aim of the piece of writing that affect its unity; such digressions may or may not trigger errors in the logical structure of the text (there can be digressions within well-structured paragraphs and texts)

Six

• Function accomplishment – captures problems that seem to be caused by the writer’s insufficient ability to manipulate the functions of language; the writer’s intended meaning is, in such situations, assumed based on the task assignment or on the aims stated by the writer themselves; e.g. the learner is supposed to write an argumentative essay, but he/she merely produces descriptive text; this problem may be discernible with regard to particular passages within the text or the text as a whole

• Clarity of expression – concerns situations when the writer fails to make themselves clear within the particular context of the text; this category thus targets unclear messages that cannot be assigned to error in grammar or lexis (can be combined with unnatural language, however) and are ambiguous, nonsensical, or too vague; one of the causes might be wordiness and long complex sentence structures

Seven

• Register – covers errors connected with inappropriate choice of linguistic forms with regard to the context of situation, more precisely the field of discourse (the subject matter of the language use), the tenor of discourse (the relationship between participants) and the mode of discourse (written as opposed to spoken); often involves errors connected with formality and the stylistic meaning of lexis; connected with genre appropriacy

• Function appropriacy – concerns errors of performing inappropriate functions or omitting appropriate/expected functions within the particular situational context; close connection with genre and real-world knowledge

• Reader sensitivity – sensitivity to the reader’s needs; errors in this category are primarily manifested by inaccessibility of the text to the target reader caused by wrong assessment of the shared knowledge between the author and their audience