Indefinites in English and Spanish Child Language

1
Indefinites in English and Spanish Child Language
MSU Workshop First Language Acquisition
Crosslinguistic and Intralinguistic Variation
April 3, 2004.

• Karen Miller
• Michigan State University

2
Motivation

• 1. Present a series of experiments on child
  interpretation of Spanish and English indefinite
  noun phrases and Spanish bare singulars.
• 1.1 child ability to distinguish between
  singular indefinites (ambiguous) and
  bare nominals (obligatory narrow scope).
• 1.2 the role that the lexical nature of the
  indefinite article ( / ? 'one') plays a role
  in child interpretation of indefinites.
• a. Peter didnt bring a dog.
• Pedro no trajo un perro.
• b. Peter didnt bring one dog.
• Pedro no trajo uno perro.
• 1.3 the role that discourse plays in the
  ability of children to access the wide scope
  reading of indefinites under negation.

3
2. Types of sentences that were tested. a. El
niñito no trajo una pelota The boy
neg brought a ball The boy didnt bring
any balls (neguna pelota) There was a ball
that the ball didnt bring (una
pelotaneg) The number of balls that the boy
didnt bring is one (number) b. El niñito
no trajo pelota The boy neg brought
ball The boy didnt bring any balls
(negpelota) c. Peter didnt close a
drawer. Peter didnt close any drawers
(nega drawer) There was a drawer that Peter
didnt close (a drawerneg)
4
Issues on the acquisition of scope properties of
indefinites

• 3. Empirical Issues a. Childrens scope
  preferences
• b. Cross-linguistic differences

4. Theoretical Issues a. Acquisition models
5
What we know about child preferences

• 5. Musolino, 1998 (2 of 4 guys were found)
• The detective didnt find some guys
• someneg 35 (310-66, mean age 51)
• 6. Su, 2001
• The boy didnt ride a dog (3 of 4 dogs were
  ridden)
• aneg 33 (310-62, mean age 50)
• 7. Lidz Musolino, 2002
• Donald didnt find two guys (2 of 4 guys were
  found)
• twoneg 33 (311-411, mean age 44)

6
What we know about languages other than English

• 8. Lidz Musolino, 2002 Kannada (2 of 4 cars
  were washed)
• Anoop eradu kaaru toley-al-illa
• Anoop didnt wash two cars
• twoneg 23 (40-411, mean age 45)
• 9. Kramer, 2000 Dutch (2 of 3 fish were caught)
• De jongen heeft een vis niet gevangen
• the boy has a fish not caught
• They boy did not catch a fish
• aneg 16 (42-76, mean age 58)

7

• 10. Su, 2001 Chinese (3 of 4 dogs were ridden)
• Milaoshu mei you qi yi-zhi gou
• Mickey Mouse NEG have ride one-CL dog
• Mickey Mouse didnt ride one dog.
• number reading 77 (46-63, mean age 54)

8
Proposals to account for childrens behavior

• 11. Observation of Isomorphism (Musolino,
  1998)
• Child interpretation of indefinites under
  negation correlates with the interpretation
  determined by syntactic scope. This proposal
  predicts that children will not allow the wide
  scope reading of indefinites that occur lower
  than negation in the surface syntax.
• Accounts for English, Kannada
• Does not account for Dutch
• Problems Children prefer wide-scope reading of
  overt partitives (Gualmini Musolino, 2003).

9

• 12. Non-Integration Hypothesis (Kramer)
• The specific reading of indefinites is acquired
  later (after age 7) because it requires discourse
  integration and children are unable to integrate
  discourse at this age.
• Accounts for Dutch, English, Kannada
• ProblemsStudies are showing that children are
  sensitive discourse by this age (Wijnen and
  Roeper, 2003).

10
Spanish Bare Singulars and Indefinite
Singulars a. Properties of Bare Singulars vs.
Indefinites b. Do children distinguish
between bare singulars and singular
indefinites? c. Compare Spanish child
interpretations cross- linguistically Englis
h Indefinites a. Discourse conditions for
specific reading b. Provide alternative
account of previous findings on indefinites in
English c. Present experiments that adjust
materials according to that account.
11
Spanish Bare Singulars and Indefinites

• 13. Carlson (1977) Bare plurals are not the
  plural counterpart of singular indefinites.
• 14. Bare plurals/Indefinites under Intensional
  Verbs
• a. John is looking for students narrow/wide
• J. is looking for any students
• b. John is looking for a student narrow/wide
• J. is looking for any students
• There is a student that J. is looking for
• 15. Bare plurals/Indefinites under Negation
• a. The boy didnt eat french fries narrow/wide
• b. The boy didnt eat a french fry narrow/wide

12

• 16. Distribution of Spanish bare singulars and
  indefinites (Bosque, 1996) Spanish bare
  singulars behave like bare plurals, except their
  distribution is much more restricted.
• 17. Bare singulars/Indefinites under Intensional
  Verbs
• a. Juan está buscando casa narrow/wide
• Juan is looking for house
• J. is looking for any house
• b. Juan está buscando una casa narrow/wide
• Juan is looking for a house
• J. is looking for any house
• There is a house that J. is looking for

13
18. Bare singulars/Indefinites under
Negation a. El chico no trajo
pelota narrow/wide The boy neg brought
ball The boy didnt bring any balls b. El
chico no trajo una pelota narrow/wide T
he boy neg brought a ball The boy didnt
bring any balls The boy didnt bring one of
the balls
14

• 19. Spanish bare singulars are restricted to
  complement position and are licensed under
  negation, intensional verbs, and in questions all
  of which deal with possession or lack of
  possession.
• a. Ando buscando casa g. Me compré auto
• Im looking for a house I bought a car
• b. Casa es buena tener h. Me conseguí perro
• House is a good thing to have I got myself
  a dog
• c. Quiero leer revista I. No trajo pelota
• I want to read a book He didnt bring a
  ball
• d. Me falta cuchara j. Quiere usar
  computador
• Im missing a spoon He wants to use a
  computer
• e. Hay secretaria? k. No tengo camisa limpia
• Is there a secretary? I dont have a
  clean shirt
• f. Tengo casa en Puerto Natales i. Juan
  necesitaba ayudante
• I have house in Puerto Natales Juan needed
  an assistant

15

• Non-specific indefinites across languages
• 20. Matthewson (1999) Salish polarity
  indefinites (-ku lacking number features)
  licensed by modals, negation, and questions
  non-specific reading.
• 21. Farkas de Swart (2003) Hungarian PredOp
  bare nominals licensed by negation and
  progressive aspect non-specific reading.
• 22. Van Geenhoven (1998) West Greenlandic
  incorporating nouns incorporating verbs get,
  buy, have, look for, drink, eat non-specific
  reading.
• 23. Pérez-L Roeper (1999) English bare
  singulars Inherent Binding Constructions
  (possession).
• 24. Neidle (1988) Russian DO nouns with genitive
  case licensed by negation, verbs achieve,
  attain, want, look for, demand, wish
  non-specific reading.

16
Relevant Properties for the Present Study

• 25. Bare singulars are scopeless indefinites are
  ambiguous.
• a. El niñito no trajo una pelota
• The boy neg brought a ball
• The boy didnt bring any balls
• There was a ball that the boy didnt bring
• The number of balls that the boy didnt bring
  is one
• b. El niñito no trajo pelota
• The boy neg brought ball
• The boy didnt bring any balls

17
Experiment 1

• 26. Subjects 24 Chilean children (45-511,
  mean age 50) 30 Chilean undergraduates
• 27. Task Truth Value Judgment Task
• 28. Experimental Design Within Subjects
• 4 bare singular targets
• 4 indefinite targets
• 12 fillers and controls
• 20 experimental stories

18
29. Test Sentences
19
Este niño está pensando en llevar algo al jardín
para jugar con sus amigos. Primero, pensó en
llevar sus pelotas pero después decidió llevar
sus peluches. This boy is thinking about taking
something to school to share with his friends.
First he thought about taking his balls but he
thought hed rather take his teddy bears instead.
Así es que el niño fue a buscar sus peluches
pero se acordó que a sus amigos les gusta jugar a
la pelota. Así es que decidió no llevar ningún
peluche (plausible dissent). So he went to find
his teddy bears and then he remembered that his
friends really liked playing soccer. So he
decided not to bring any teddy bears.
Al final el niño decidió traer sus pelotas al
jardín. El niño trajo la pelota amarilla, la
pelota azul y la pelota verde, pero no trajo la
pelota naranja porque estaba desinflada. So
finally the boy decided to bring his balls to
school. He brought the yellow ball, the blue ball
and the green ball but he didnt bring the orange
ball because it was flat
20
El chico no trajo una pelota The boy
neg brought a ball The boy didnt bring a
ball
FALSE (neg a ball)
TRUE (a ball neg) (number reading)
21
b. El chico no trajo pelota The boy
neg brought ball The boy didnt bring a ball
FALSE (neg a ball)
TRUE (a ball neg) (number reading)
22
30. Results
FALSE (nega)
Fig. 1. Percentage of FALSE (indefinite
narrow-scope) responses. Note A FALSE response
for the bare singular condition is the only
correct response, while the indefinite condition
is ambiguous so an answer of either FALSE or TRUE
is available in the adult language depending on
the interpretation that is accessed.
23

• 30.1 The proportion of FALSE responses for
  each child was entered into a 2 (Age adults,
  4-5 year old children) X 2 (Condition bare
  singulars, singular indefinites) mixed design
  Analysis of Variance (ANOVA) with Condition as a
  within subjects variable and Age as a between
  subjects variable. The analysis revealed a main
  effect for Age (F(1,52) 4.388, p main effect for Condition (F(1,52) 61.781, p .0005). However, there was no significant
  interaction between Age and Condition (F(1,52)
  1.641, p .206.
• 30.2 Main effect for CONDITION children like
  adult distinguish
• between indefinites and bare singulars.
• 30.3 Main effect for AGE although children do
  not reach adult levels on both tasks, there is
  no question that they treat indefinites and bare
  singulars differently and follow the overall
  trend found for adults.

24

• 31. Given very similar experiments, Chinese
  children and Spanish children pattern with adults
  on their responses, while English children and
  Dutch children do not.
• 32. Chinese experimental sentences involved
  numeral yi-ge (one) English and Dutch
  experimental sentences involved indefinite
  articles.
• 33. Spanish indefinite article is ambiguous
  between numeral and indefinite article. Spanish
  children appear to be accessing a number reading.
  

Table 1. Rejection of Target Sentences in
Indefinite Singular Condition
indicates significant difference between adult
and child.
25
English Indefinites

• Commonality among previous studies They all
  deal with covert partitives
• a. M.M. didnt ride a dog
• (Context M.M. rode 3 of 4 dogs)
• M.M. didnt ride one of the dogs
• b. The detective didnt find two (of the)
  guys
• c. The boy didnt catch a (one of the) fish
  
  

26
Working assumptions about partitives

• 35. Enç (1991) indefinites become specific when
  linked to discourse one way of linking a
  variable to prior discourse is through a subset
  relation (a relation of inclusion), as in
  partitive constructions.

36. Geurts (2002) indefinites always denote
properties if an indefinite occurs as an
argument it may be construed as specific or
non- specific depending on whether it is
backgrounded or not.
27
37. Partitives and Backgrounding (Geurts,
2002) -given an expression of the form det ?
of ? -? job is to help identify the intended
? -hence ? is backgrounded

• Mary didnt eat a piece of her toast
• ? a piece
• ? her toast
• b. Mary didnt close a drawer (of the dresser)
• ? a drawer
• ? the drawers that belong to a dresser that
  was previously mentioned in discourse

28
Alternative explanation for the previous results
38. Backgrounding Hypothesis Children have
difficulty backgrounding ? in covert partitives
because ? is covert, not spoken. If ? is not
backgrounded then ? cannot be interpreted
specifically.

• a. The boy didnt catch a fish (fish in a pond)
• b. Mickey Mouse didnt ride a dog (dogs in a
  circus)
• Donald didnt find some guys (guys in the
  forest?)
• The girl didnt buy a shirt (shirts in a store)
• e. Anoop didnt wash two cars (cars at his work?)

29

• Evidence from Previous Studies

39. Kramer (1998) follow-up experiment
introduced the relevant source set (i.e. group
of fish) at the beginning of the story. (age 4
710) Specific readings increased slightly
from 16 to 27, but mainly with older children.

• 40. Gualmini Musolino (2003) (44 age)
• Overt Partitives 73 wide-scope
• Covert Partitives 25 wide-scope

30

• 41. Can making the source set more natural
  (providing an entity that connects all the
  members of the set together) will help children
  background the relevant set against which they
  can subsequently interpret the indefinite
  specifically?
• candles on a birthday cake
• drawers in a dresser
• letters on a chalkboard
• eggs in a basket

31
Experiment 1A

• 45. Subjects 20 children (310-55, mean 42)
• 10 undergraduates
• 46. Task Truth Value Judgment
• 4 Targets
• 4 Controls
• 12 Fillers

32

• 47. Test Sentences
• a. Mary didnt paint an egg (? egg ?
  eggs in basket)

b. Timmy didnt close a drawer (? drawer
? drawers in dresser)
33

• Peter didnt blow out a candle (? candle
• ? candles on cake)

• Suzy didnt erase a letter (? letter
• ? letters on chalkboard)

34
Researcher This is Peter and these are drawers
that are all in his dresser and hes
closing them. Peters dad told him to
close all of the drawers before going
outside to play. Lets see what happens.
35
Researcher Look. Now Peter is going outside to
play. But wait! What about this one? He
didnt close this one (pointing to open
drawer). Lets see if the monster can say what
happened.
36
48. Peter didnt close a drawer
37

• 49.. Results Overall Percentage of Wide-scope
  Readings
• Children 91
• Adults 100
• 75 of children gave wide-scope response 100
  of the time.

38
Experiment 1B (follow up)

• 50. Subjects 13 children (42-58, mean 49)
• 14 undergraduates
• 51. Task Truth Value Judgment
• 52. Test Sentences 1. Mary didnt paint an egg
• 2. Jonathon didnt erase a letter
• 3. Tommy didnt close a drawer
• 4. Pam didnt blow out a candle

39
53. Researcher This is Peter and these are
drawers that are all in his dresser and hes
closing them. Peters dad told him to
close all of the drawers before going
outside to play. Lets see what happens.
40
54. Researcher Look. Now Peter is going outside
to play. But wait! What about this one?
Lets see if the monster can say what
happened. (omitted He didnt close this one.)

41

• 55. Results Overall Percentage of Wide-scope
  Readings
• Children 92
• Adults 96
• 85 of children gave wide-scope response
  100 of the time.

42
Experiment 2

• 56. Subjects 11 children (47-75, mean 47)
• 20 undergraduates
• 57. Task Truth Value Judgment Story acted out
  with dolls.
• 58. Test Sentences 1. Mary didnt paint an egg
• 2. Jonathon didnt erase a letter
• 3. Tommy didnt close a drawer
• 4. Pam didnt blow out a candle

43
Researcher This is Timothy and these are letters
that are all written on the chalkboard and hes
erasing them. His teacher told him to erase all
of the letters before going out to recess. Lets
see what happens!
44
(Timothy erases first letter)
45
(Timothy erases second letter)
46
(Timothy erases third letter)
47
Timothy Hmmm, Im tired. Im going out to
recess! Researcher But wait! Hes not finished!
Lets see if Petey (the puppet) can say what
happened.
48
59. Timothy didnt erase a letter.
TRUE aneg
FALSE nega
49

• 60. Results Percentage of Wide-scope Readings
• Children 77
• Adults 89
• 64 of children gave wide-scope response
  100 of the time.

50
Summary of Results
(62) By 310 children do not follow a surface
scope principle for interpreting indefinites but
instead appear to have difficulty integrating
discourse (backgrounding the relevant set for
interpreting the indefinite).
51

• Geurts (2002) Indefinites by default are
  non-specific because they tend to carry new
  information.
• It is only under special circumstances that
  new information is backgrounded and in this case
  the indefinite in question can be interpreted
  specifically.

52

• 63. Partitives and Backgrounding
• -given an expression of the form det ? of ?
• -? job is to help identify the intended ?
• -hence ? is backgrounded

64. In covert partitives, ? is covert. a. Timot
hy didnt close a drawer ? a drawer ?
the dresser
65. In sentences like Timothy didnt close a
drawer, there is one dresser that all of the
drawers are a part of. Hence the dresser defines
the set of drawers. Children are able to
background the dresser and use it to interpret a
drawer specifically.
53

• 66. In previous studies, indefinite NPs were
  interpreted non-specifically because children
  were unable to background the set against which
  the indefinite was supposed to be interpreted.

• dogs at a circus
• fish in a pond
• guys in a forest

• candles on a birthday cake
• letters in a chalkboard
• eggs in a basket

54
Conclusions

• Several factors influence how children interpret
  sentences involving negation and indefinites
• a. The lexical nature of the indefinite
  (whether it is also the word for one) appears
  to allow an additional reading for Spanish
  children that is not available for English and
  Dutch children. It appears that Spanish children
  are accessing a number reading of the indefinite
  in addition to the narrow scope reading.
• b. It appears that wide-scope readings for
  English-speaking children are not impossible
  when discourse conditions are met. Instead, when
  children can background the relevant set against
  which to interpret the indefinite,
  English-speaking children as young as 4 years of
  age appear to have no trouble accessing the wide
  scope reading of indefinites under negation.

55
References Bosque (1996) Por qué determinados
sustantivos no son sustantivos determinados.
Repaso y balance. in El sustantivo sin
determinación. La ausencia de determinante en la
lengua española, ed. Ignacio Bosque. Visor
Libros, 13- 121. Carlson, G. (1977) A unified
analysis of the English bare plural. Linguistics
and Philosophy, 1, 413-457. Farkas, D. H. de
Swart (2003) The semantics of incorporation. CSLI
Publications. Lidz, J. and J. Musolino (2002)
Childrens command of quantification, Cognition
84, 113-154. Krämer, I. (2000) Interpreting
Indefinites An experimental study of childrens
language comprehension, Ph.D. Dissertation,
University of Utrecht. Masullo, P. (1992)
Incorporation and Case theory in Spanish a
crosslinguistic perspective, Ph.D. dissertation,
University of Washington. Matthewson, L. (1999)
On the Interpretation of Wide-scope Indefinites.
Natural Language Semantics, 779-134.
56
Miller, K. and C. Schmitt (2003) Wide-scope
Indefinites in English Child Language. Paper
presented at GALA2003, Utrecht, The
Netherlands. Musolino, J. (1998) UNIVERSAL
GRAMMAR AND THE ACQUISITION OF SEMANTIC
KNOWLEDGE an experimental investigation into the
acquisition of quantifier-negation interaction
in English. Ph.D. dissertation, U of
Maryland. Neidle, C. (1988) The Role of Case in
Russian Syntax. Kluwer Academic
Publishers. Pérez-Leroux, A-T. and T. Roeper
(1999) Scope and the Structure of Bare
Nominals Evidence from Child Language,
Linguistics 37, 5, 927-960. Su, Y.-C. (2001)
Scope and Specificity in Child Language In Ann
H.-J. Do et al. (eds.) BUCLD 25 Proceedings,
744-755. Somerville, MA Cascadilla Press. Van
Geenhoven, V. (1998) Semantic Incorporation and
Indefinite Descriptions. CSLI Publications,
Stanford, California.