A linguistic representation is a representation of a language sign or of an aspect of it in some visual medium, typically in writing. The representation is linguistic in the sense of ‘based on a theory and a method of linguistics’. As always, the language sign may be of any complexity; a whole discourse, conversation or text are included.

The methodological status of the piece of text represented, viz. as a piece of data (constituting linguistic evidence; Lehmann 2004) or as an example (illustrating some statement), may make a difference for the style of representation.

Linguistic representations occur chiefly in the following contexts:

  1. edition of a text
  2. example inserted in a metalinguistic context
    • embedded in the running text
    • formatted as a separate block.

Levels of representation

The two fundamental aspects of the language sign are its significans and its significatum. They found distinct semiotic levels and different methodological levels of representation. Leaving aside combinations of representations (s. below), any single representation can only render one of them. However, by convention, the sign as a whole may be represented by its significans, since this is its perceptible side and the representation is a perceptible object, too. This generates a potential ambiguity, as a given representation might mean the whole sign or only its significans. This distinction is made by notational conventions, especially by the types of brackets shown in the following table.

Linguistic representations
semiotic levelmethodological levelaspectexample
1significansphoneticsauditory: form of utterance tokens[ˈgnɔ́ɔtʰisauˈtɔ́n]
2phonologyauditory: systematic form/gnôːtʰisawtón/
3orthographywritten: speech community formΓνῶθισεαυτόν.
4transcriptionwritten: form in metalanguage<Gnṓthiseautón!>
5signmorphophonemicsmorphological structure{gnôː-tʰise=awt-ón}
6morphologymeaning of
7syntaxsyntactic structure[[[gnôː-tʰi]V[se=awt-ón]NP]VP]S
8significatumsemanticstranslation in metalanguageKnow thyself!

This enumeration of methodological levels is far from complete: a prosodic, a discourse representation, a semantic representation and several others could be mentioned in addition.

Linguistic representations are embedded in a metalinguistic context. The following are worth mentioning:

Combination of representations

Some linguistic representations are more than one-dimensional. For instance, a tree diagram is two-dimensional, and so is a discourse representation of conversation analysis. All of the representations in the above table are one-dimensional (disregarding diacritics). Several of them may combine in a multilineal representation of one piece of text, essentially as displayed in the table.

A multilineal linguistic representation does take the form of a table (Lieb & Drude 2000). The table rows are constituted by different representation levels. The columns usually separate word forms. This typically holds for representations which display grammatical properties of the piece of text, including a morphological glossing. In such cases, a word-by-word alignment of the lines applies to levels #5 - 7 of the table. Two of the levels are normally exempt from alignment by such grammatical units, viz. #1 and 8.

The order of the lines in a multilineal representation is generally from significans to significatum, again as in the above table. In particular, levels #5, 6 and 8 come in this order.


Lehmann, Christian 2004, “Data in linguistics.” The Linguistic Review 21(3/4):275-310. [download ]..

Lieb, Hans-Heinrich & Drude, Sebastian 2000, Advanced glossing: A language documentation format. Berlin: Technische Universität (Working Papers).