Dowty's (1979) classified predicates into:

1. States 'the woord is burning' ---Atelic
2. Activities 'Mary pushed the cart' ---Atelic
3. Accomplishments 'Mary melted the chocolate' ---Telic 
4. Achievements 'Mary crossed the finish line' ---Telic

My question in relation to Dowty's classification is how the atelic/telic diagnostics are configured, are there tests? Or simply they are based on aspect? I believe that the answer to the second question if indeed this is based on aspect properties (Aktionsart), it would not reveal why sometimes even during the Perfective/past [Finished Action] some predicates are atelic like in 2. I have read in Borer (2005b) 'the normal course of events' on page 52, that there are some 'truth conditional tests' used to determine atelicity/telicity. But still the picture is always blurry for me. Could anyone here explain this please. Thank you.

NB: I'm not sure about the atelicity of (1), I was only assuming.


Just found my old notes from grad school. It's not quite clear if your interest in telicity is historical (e.g. how did Dowty 19XX understand telicity?); obviously, a lot has been done since then.

The following is based on Depraetere 1996.

Tests to distinguish boundedness from unboundedness and telicity from atelicity

(15a) John drank beer. (atelic, unbounded)

(15b) John drank a glass of beer. (telic, bounded)

Test 1.

Bounded – “how long does it take X to X?”

Unbounded – “for how long?”

How long did it take John to drink a glass of beer? John drank a glass of beer in half an hour. BOUNDED

For how long did John drink beer? John drank beer for half an hour. (caveat: the answer becomes bounded due to a for-adverbial insertion). UNBOUNDED

“A non-progressive atelic sentence does not allow the use of an in-adverbial (17a), whereas a non-progressive telic sentence does (17b):

(17a) I lived with her for two years/*in a year.

(17b) I did the exercise in 10 minutes.

Limitations of the test:

a. certain sentences can collocate with a for-adverbial as well as with an in-adverbial (zero-telic verbs):

(18a) I mowed the lawn in half an hour. TELIC

(18b) I mowed the lawn for half an hour, then my brother took over. ATELIC

b. the test can only be used for (a)telicity if the sentence is non-progressive, as progressive telic sentences do not collocate with an in-adverbial:

(19a) ?? He was pushing the cart into the barn in 15 minutes.

A for-adverbial is acceptable in this context but it has the effect of turning a telic sentence into an atelic one (cf. example 18b):

(19b) He was pushing the cart into the barn for 15 minutes.

c. might be problematic with non-progressive telic punctual situations; use ‘in one second’ instead:

(20a). The bomb exploded in one second.

d. not useful in sentences already containing a duration adverbial;


Unbounded – homogeneous situations, i.e. the situation remains the same throughout the time it holds. (subinterval property)

Bounded – heterogeneous situations, “they cannot be broken down into smaller entities that can be indicated with the same name” (Declerck 1991: 264)

(15a) John drank beer. (atelic, unbounded)

(15b) John drank a glass of beer. (telic, bounded)

Note: applicable to (a)telic propositions only and not actual progressive sentences.


Stop+V-ing (atelic)

Finish+V-ing (telic)

John stopped drinking beer. (atelic)

John finished drinking a glass of beer. (telic)


If one was verb-ing but was interrupted while verb-ing, has one verb-ed? If the answer is possible – atelic.

If John was drinking beer and he stopped drinking beer, has he drunk beer? Yes. (atelic)

If John was drinking a glass of beer and he stopped drinking a glass of beer, has he drunk a glass of beer? Not necessarily. (telic)

Depraetere, I. (1996). The tense system in English relative clauses: A corpus-based analysis. Berlin; New York: Mouton de Gruyter.

| improve this answer | |
  • Dear Alex, thank you for this useful and summarized answer. Now at least I know how the logic works (though still need to learn a lot). If you don't mind I have another last question. Since we consider atelicity & telicity (or unboundedness & boundedness) as a property of Aktionsart (Lexical aspect) that a verb or predicate determines, can we say that these properties don't vary according to voice alternation? Or do they indeed vary? thank you again! – Tsutsu May 1 '18 at 9:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.