Movement Constructions - Passives

Movement Constructions - Passives

1. Generativist vs. constructivist

   • Good test case as movement vs. learnt construct.
   • English passive takes form of subject - verb - object. The patient is before the agent compared to agent before patient in actives.
   • Generativist: argument movement rule where start with a statement then move patient to the front of the verb. Born with it so all children should be good at passives but aren't; better with actional passives.
   • Constructivist: passives are learned gradually from the input. Learn holophrases first, then slot and frame patterns then abstract constructions.

2. Generativist

   • Borer & Wexler (1987) the argument movement rule is innate but matures at 5 years old.
     • Doesn't explain why they are good at other passives before 5 but claim that they aren't real passives they are just adjectives.
     • No longer use the A movement rule for generativists - believe children are good at passives it just seems like they aren't.
   • 4 reasons for why children appear to be bad with passives:
     • Familiar verbs: verbs that children are used to hearing in actives are harding to comprehend and produce in passives.
       • Pinker, Lebeaux & Frost (1987) children were much better with passives like the boy was glorped by the girl than passives with familiar verbs which they didn't beat chance until 5.
     • The by-phrase: children find passives difficult when there is failure to include the by-phrase.
       • Crain & Fodor (1993) two soldiers hit by either hulk or darth vadar. When child produced passives they were much better when including the by-phrase such as the soldier was hit by darth vadar.
     • Difficultly illustrating: children just appear to be bad at non-actional passives because it is hard to illustrate these in pictures that are used in the studies.
       • Messenger et al (2012) 3-4 year olds were bad at all experience theme pictures (e.g. fear) for both actives and passives.
     • Reversible passives: most studies use reversible passives where either person could be doing the thing.
       • Bencini & Valian (2008) in real life we experience more irreversible passives and even 3 year olds are able to produce full passives using this method e.g. the lemon was cut by the knife.

3. Constructivist

   • Israel, Johnson & Brooks (2000) naturalistic at 2:6, 2:7 and 3:7. Stative, then equivocal then eventive passives indicating learning but this would look similar for adults because some are more common than others so need experimental data.
   • Brooks & Tomasello (1999) train children with novel verb passives. 80% at 3 and 90% at 4 suggesting they are learning from the input.
   • Savage et al (2004) low lexical and high lexical overlap passives. Children better at producing after high lexical overlap and only older children good at low lexical overlap. Supports stages of learning.
   • Ambridge et al (2015) passives where subject has something done to it by object and is highly affected. Hearing a lot of these should make child better at them. Adults rated more effective verb passives as better sentences and had faster reaction times.
   • Problems; children should have this verb by verb difference but don't and little research on lexical effects.