1 Among the differences between spoken and written discourse that are exemplified in this extract are:

– Lack of clear sentence definition: while most written text consists of clearly defined sentences, beginning with a capital letter and ending with a full stop, speech is far less clearly segmented. For example:

Just knew she was about my age and what she looked like just from looking at her and and uh so I asked my friends about her like what they knew about her and stuff.

For this reason, it is often easier to analyse it in terms of tone units, utterances or speakers’ turns.

– Lack of clause complexity: utterances are often strung together and joined with co-ordinating conjunctions (and, but, so, etc.), rather than showing the kind of internal complexity typical of much written language, where subordinate clauses are often frequent.

– Lack of phrase complexity: similarly, phrases tend to be shorter on average and less complex.

– Vague language: e.g. what they knew about her and stuff; this place; and go to like a club. Since speaking usually takes place in real time, there is a lot of pressure on speakers to get their message across without too much time spent searching for the most accurate term or expression.

– Repetition and re-phrasing (in a near town town slash city); filled pauses (And um so); false starts and self-repair (we’d go we’d hang out) and grammatical inaccuracy (so I can go see her). Again, these are effects of real-time planning pressure.

– Ellipsis, i.e. the omission of elements that can be ‘recovered’ from the context: [I] Just knew she was about my age.

– Formulaic language (see below): never said a word; did my homework; dance the night away.

An obvious difference (and not apparent in this transcript) between speaking and writing is, of course, the use of prosodic devices such as stress, rhythm and intonation.

2 right: this is an all-purpose question tag; tags are added at the end of a clause and typically ask the listener for confirmation – sometimes called a ‘response getter’.

Cool: an evaluative ‘insert’: inserts are stand-alone elements that are common in spoken language and have a variety of functions.