Tema 8 – General structure of lexical

Tema 8 – General structure of lexical entries. The content of lexeme 1: phonological information.

Lexical entry = lexical entries in the mind (representation of lexical meaning in the speakers’ minds, part of their competence)

Desk dictionaries provide the following information in entries:

• **Word form (**spelling)
• Pronunciation
• Syntactic category / part of speech
• Definitions
• Syntactic context
• Quotations / examples of usage
• **Related inflectional /derivation also forms (**drink/drinker)
• Pragmatic labels: colloquial, dialectal, etc. (the context the word is used in, formal, informal)
• Etymology
• Encyclopaedic / general information

However, not all of these are available to native speakers necessarily. For example, we have seen that etymological information is not available to most speakers.

Similarly, the general information we have on lexical concepts may vary significantly across individuals. Therefore, our ideal speaker would have this basic template: (Idealisation who is competence in the use pf language)

• Word form (spelling)
• Pronunciation (PHON)
• Definitions (SEM)
• Syntactic category / part of speech (SYN)
• Syntactic context (SYN)
• Related inflectional forms
• Related derived forms (drink/drinker) (MORPH)

In other words, a competent speaker should have at least PHON, SEM, SYN and MORPH information for each lexical item (in your mind) of his/her language.

Our next step is to represent the speaker’s lexical competence. But, how we do that?

• We will construct lexical entries which contain at least that information.

How do we construct our lexical entries?

There are many different ways to state lexical knowledge, but we would like our system of representation to be:

• 1. Explicit (system don’t need to …. representations)
  2. Simple (because what we want to do is to represent linguistic information, so we are not interested in the system itself. Not relevant for us. We are interested in linguistic information, tool to state linguistic information. Seen as a tool not as a ..)
  3. Faithful to the organisation of ML (compatible with what we know or store in the lexical mind)

And san an additional “bonus”

• 1. Computer tractable (you can use it to construct representation which can manipulate the way we produce the language)

This is what we find in the formalism known as:

[ATTRIBUTE: VALUE] or [A: V]

The reasoning behind it is very simple: we define a number of attributes for a lexical entry as an initial template(plantilla), and we then assign a value for the lexeme at issue

(You will know which attribute must provide with values)

Hay una serie de atributos a los que le tengo que dar un valor. General = atributos -> que a su vez debemos darle un valor a ese atributo (el nombre). Vamos dando valores a atributos definidos. hAy tambien plantillas.

Forum type. Para este atributo puede ser simple o limitarlo (ejemplo una participación por estudiante). Atributos están definidos y yo decido los valores.

Los atributos estan definidos y yo tengo que decidir los valores. Cuando hablo yo si utilizo un verbo decido en que forma verbal lo pongo, pero a partir de unos atributos (el lenguaje te limita, te da un atributo. Para el cual tu tienes que dar un valor).

Let us see an example:

The lexeme “organise” will contain the following phonological information (PHON), where the Value for the Attribute “PHON” is a simple phoneme transcription.

Attribute is given -> we have this attribute but we need to provide one value or another because they have different pronunciation. However, we could be more specific in the PHON box and create further (A: V) pairs with in:

This complex combination of (A: V) pairs is knows as a Feature Structure.

**Always the same idea/system for the same attribute but you can make it complex.

This gives us the following representation for “organize”

Note that we need not repeat the value ): for syl (1) in stress”

The complete template for lexical entries is given in the handout.You will see that this is the maximum template. Some attributes may be irrelevant for many lexemes.

CAT = attribute. PERSON- NUMBER- GENDER… = all of that are attributes too.

Template given to you and you provide values in the lexeme you are describing . For example a verb you have to provide the form verb , the paste in that case.

*Example: attribute = give up -> you have to do values for those attributes ?

**Estamos construyendo entradas lexicas para el now mental.