Chapter 10Interlinear Glossing, Abbreviations, and Citation Conventions

10.1 Reading an example

An interlinear example presents an Ainu expression, an aligned gloss, and an English translation. The gloss identifies the parts of the expression; the translation conveys its meaning in context. The format follows the Leipzig Glossing Rules Comrie, Haspelmath & Bickel (2015), with a separate convention for marking person forms explained below.

The first two words are glossed with English lexical equivalents. In the verb, ku= identifies a first-person singular subject: 1SG.A=give reads “first-person singular transitive subject + give.” The past-tense form “gave” belongs to the English translation; no separate past-tense marker appears in the Ainu example.

Some examples also show the unsegmented pronunciation, the source's katakana spelling, or a literal translation. These additions help where a morphological division is difficult to recover from the pronounced form, or where an idiomatic translation conceals a useful structural parallel.

10.2 Boundaries and combined labels

Hyphens and equals signs in the Ainu line are matched in the gloss. A dot occurs within a gloss: it lets several labels describe one form without dividing that form into smaller pieces Comrie, Haspelmath & Bickel (2015).

The equals sign follows Nakagawa's convention for making person markers visible while leaving the stem easy to identify. He explains that it originated as a way of distinguishing person marking from derivation Nakagawa (2024: 59–61). In the Leipzig rules, the same sign marks a clitic boundary; its use here therefore needs this qualification. Whether a particular person marker behaves as an affix or a clitic is a grammatical question treated in Chapter 69 (Person-Marker Order and Stem Boundaries).

10.3 Grammatical abbreviations

Lexical meanings appear in lower case, such as give and children. Grammatical labels appear in capitals or small capitals. The complete list is in Chapter 174 (Abbreviations and Glossing Conventions). Frequently used labels include the following:

The labels S, A, and O identify argument roles without assuming that Ainu treats all subjects alike Bugaeva (2012: 471). Likewise, 4 identifies a series of forms whose interpretation varies with the construction. Its uses include indefinite reference and person reference in narration; see Chapter 67 (Fourth-Person Forms and Their Uses). A gloss is a compact analysis of a form. The accompanying discussion explains ambiguities that a short label cannot express.

10.4 Dialect and speaker attribution

Dialect tags identify the provenance of examples. HK means Hokkaido without a more specific identification; it does not mean that a form has been documented throughout Hokkaido. Locality and speaker information, when available, appears separately.

An attested example cites the work or recording in which it occurs. An example composed to illustrate a pattern carries the label constructed example. Such an illustration cannot establish that a speaker used that sentence. The limitations of dialect and corpus sampling are discussed in Chapter 9 (The Dialect Sample and the Corpus-Quantification Method).

10.5 Following a citation

Citations give the author, publication year, and, where relevant, a page, section, or example number. Opening a citation shows the bibliographic entry. The references at the end of each chapter list its cited works; Chapter 175 (References and Bibliography) collects the bibliography for the grammar.

A reported badge identifies a work cited through another source. Where an example comes from such a quotation, its note identifies the intermediary. This matters when checking spelling, glosses, or interpretation: a later author's transcription may differ from the original edition. Adapted boundaries, glosses, and translations are identified in example notes when needed to understand the relationship to the source.

10.6 Evidence labels

A label in angle brackets at the end of a claim describes its evidential status. Ordinary unmarked exposition presents established descriptions. The labels distinguish disagreement, findings from a specified corpus sample, and hypotheses requiring further support.

A corpus label applies within the limits of the sample described. It does not turn the absence of a form in that sample into evidence that the form is impossible in the language.