A crosslinguistic database on Tough constructions
Tough constructions are a type of sentence structure with a unique interpretational dependency. This website is a cross-linguistic database of such constructions.
Consider the following. This is not a tough construction.
1) John tried to kick the ball.
In this sentence, there are two verbs try and kick which indicates that there are two agents who carry out each action. All native speakers of English can only interpret this sentence one way: The person who is the agent of trying is the person who kicked the ball. Now consider (2).
2) John convinced Paul to kick the ball.
In this sentence, there are again two verbs convince and kick. In this sentence, the person who is the object of convince is the agent of kicking the ball.
(1) and (2) show that the subject or object of the higher verb can be interpreted as the agent of the lower verb.
The following is a variation of (1) and (2).
3) John tried to be kicked.
4) John convinced Paul to be kicked.
In these sentences, the subject and object of the two sentences respectively are interpreted as the patient of the lower verb. Notably, for this interpretation to arise, the lower clause must be in passive form.
Tough constructions are different.
5) The book was easy to read.
In this sentence, there is a subject in the higher clause just like (1), (2), (3) and (4). This subject is interpreted as the patient of the lower clause like (3) but there is no passive voice allowed in the lower clause. This is different from the previous sentences.
While English Tough constructions are the ones that are mostly discussed in the literature, such constructions also exist in related German, French and Spanish. Such constructions are also found in wholly unrelated languages like Tamil (Dravidian), Malay (Austronesian), Tagalog (Austronesian), and Tibetan (Sino-tibetan). It is fair to assume that this is quite a productive sentence type.
This website is an attempt at documenting this construction in various languages in order to answer the following (non-exhaustive list of) questions:
a) How is the interpretational dependency established in (3)?
b) What are the distinct strategies that natural languages employ to realize this construction?
c) Where does the agent interpretation of the lower verb come from?
A better understanding of these constructions cross linguistically is hoped to illuminate several questions pertaining to the syntax and semantics of natural languages.
Consider the following. This is not a tough construction.
1) John tried to kick the ball.
In this sentence, there are two verbs try and kick which indicates that there are two agents who carry out each action. All native speakers of English can only interpret this sentence one way: The person who is the agent of trying is the person who kicked the ball. Now consider (2).
2) John convinced Paul to kick the ball.
In this sentence, there are again two verbs convince and kick. In this sentence, the person who is the object of convince is the agent of kicking the ball.
(1) and (2) show that the subject or object of the higher verb can be interpreted as the agent of the lower verb.
The following is a variation of (1) and (2).
3) John tried to be kicked.
4) John convinced Paul to be kicked.
In these sentences, the subject and object of the two sentences respectively are interpreted as the patient of the lower verb. Notably, for this interpretation to arise, the lower clause must be in passive form.
Tough constructions are different.
5) The book was easy to read.
In this sentence, there is a subject in the higher clause just like (1), (2), (3) and (4). This subject is interpreted as the patient of the lower clause like (3) but there is no passive voice allowed in the lower clause. This is different from the previous sentences.
While English Tough constructions are the ones that are mostly discussed in the literature, such constructions also exist in related German, French and Spanish. Such constructions are also found in wholly unrelated languages like Tamil (Dravidian), Malay (Austronesian), Tagalog (Austronesian), and Tibetan (Sino-tibetan). It is fair to assume that this is quite a productive sentence type.
This website is an attempt at documenting this construction in various languages in order to answer the following (non-exhaustive list of) questions:
a) How is the interpretational dependency established in (3)?
b) What are the distinct strategies that natural languages employ to realize this construction?
c) Where does the agent interpretation of the lower verb come from?
A better understanding of these constructions cross linguistically is hoped to illuminate several questions pertaining to the syntax and semantics of natural languages.