View Single Post
  #58 (permalink)  
Old 18-Mar-2007, 10:01
Casiopea's Avatar
Casiopea Casiopea is offline
VIP Member
 
Join Date: Sep 2003
Country: Canada
Posts: 12,997
Current Location: China
First Language: English
Thanks: 0
Thanked 16 Times in 16 Posts
Casiopea is on a distinguished road
Default Re: The book reads well.

Quote:
Originally Posted by Andrew Whitehead View Post
If the mediopassive construct is going to carry the idea of 'stative', then the verb that carries the action in the construct has to carry the idea of 'stative' too.
Which it does. The verb's theta roles don't change here:

Active: X reads Y (Y is acted upon)
Mediopassive: Y reads well (Y is described as having been acted upon)

Now, if you're saying the verb read has to be lexically stative in order to be compatible with mediopassive voice, then given that logic, wash, a dynamic verb, shouldn't work in mediopassive voice, and yet it does; e.g., The clothes wash easily.

Quote:
Originally Posted by Andrew
With mental activities nothing happens to the object, so they shouldn't be used in mediopassive.
Why should "do" be a criterion? You need to explain that more.

Quote:
Originally Posted by Andrew
Transwicki:'a grammatical voice in which the actor of a stative verb is not expressed'
You're reading lexically stative. It's not. It's structurally stative. An intransitive verb that appears active in its morphology but expresses a passive action in its theta roles characterizes mediopassive voice; e.g., The braille book <patient> reads well.

Quote:
Originally Posted by Andrew
...your PoV is that people are using 'the book reads well' so we should ignore the current semantics of 'read' and invent a new set that matches use.
Don't do that. Don't ignore the sematics of read, ignore the urge to interpret the structural subject (e.g., The book reads well) as the agent. The book is not the agent, right? How could it be? Lexically, it's the verb's semantic object, that is, the verb read subcategorizes for a <patient> role as its object) which makes The book a <patient> and acted upon - which is how speakers interpret it (See Fagan).

The book doesn't switch its semantic role from <patient> to <agent> in passive voice, so why should it switch roles in mediopassive voice?

As a <patient> The book doesn't act, it is acted upon - no matter where in the sentence it sits. Its sematic role (as part of the verb read's subcategorizational frame) is constant; it's the syntax (i.e., the word order) that's variable. That interplay between syntax and semantics is the very reason a mediopassive verb is understood as expressing a stative meaning: The book <patient> reads well. The book cannot act. So, ignore the urge to interpret it as an <agent> that can act.

All the best.

Last edited by Casiopea; 18-Mar-2007 at 10:07.
Reply With Quote