The link between lexical semantic features and children’s comprehension of English verbal be-passives

ABSTRACT

Children seem to be relatively delayed in their comprehension of the verbal be-passive in English, compared to their acquisition of other constructions of object-movement such as wh-questions and unaccusatives. Prior work has found that children’s performance on these passives can be affected by the verb’s lexical semantics. Through a meta-analysis of experimental studies assessing English-speaking children’s age of acquisition for the verbal be- passive, we identify a developmental trajectory composed of five classes, where each class has a distinct lexical semantic profile. A Truth-Value Judgment (TVJ) Task assessing English-speaking children’s comprehension of verbal be-passives supports this developmental trajectory. Together, the meta-analysis and TVJ study underscore the importance of lexical semantics for understanding the development of the English verbal be-passive.

Acknowledgments

We would like to thank the members of the University of Connecticut Language Acquisition Group (AcquiLab), in particular, William Snyder, Diane Lillo-Martin, and Letitia Naigles, as well as Nic Schrum for lending his voice as Max Rebo the puppet. Many thanks to our adult participants at UConn and to the teachers, parents, and children at several Connecticut daycares and the Connecticut Science Center, as well as Sara Deangelo and UConn K.I.D.S. for their help in recruitment. We are additionally very grateful to Kamil Ud Deen for insightful feedback and encouragement on previous versions of this work, as well as three anonymous reviewers and the audiences at WCCFL35, the LSA Annual Meeting 2017, CLS 53, and BUCLD44 for comments and discussion about the ideas presented here.

Appendix A. Materials used in the Truth-Value Judgment Task

believe - Active

Uncle is babysitting his nephew, Luke.

Luke runs into the living room and says to Uncle.

Luke: I saw a unicorn today. It was really big and shiny. Do you believe me?

Uncle: Of course I believe you. I saw one too! Do you believe me?

Luke: No! I was the only one that saw the unicorn.

believe - Passive

Auntie is babysitting her nephew, Wyatt.

Auntie: Wyatt, I ate all of my vegetables today. Do you believe me?

Wyatt: Of course, I believe you! You love vegetables. I also ate all of my vegetables today.

Auntie: But I didn’t pack any vegetables in your lunch today! I don’t believe you.

believe - Passive

Joel and Jane are best friends.

Jane: My parents are taking me to Disney World for my birthday. Do you believe me?

Joel: I definitely believe you. Your parents are so nice. My parents forgot about my birthday.

Jane: I don’t believe you. I heard that your parents are throwing you a birthday party this year!

discover - Active

This is a story about a lion and Edward the explorer.

The lion is roaming the safari. Then, Edward the explorer arrives.

Edward: What’s this? Look, I’ve discovered a lion in the safari.

discover - Passive

The thief is hiding behind a tree with his stolen diamond ring.

Thief: Ha! I’m safe behind this tree with my ring. No one will be able to discover me in this park. I’ll take this time to sleep.

Maria is walking around the park.

Maria: Hmm … Is anyone here? Who’s this? I’ve discovered the diamond thief!

discover - Passive

This is a story about Michael and a pirate. The pirate is hiding away on an island with her treasure chest.

Pirate: Arg! I’m safe on this island with my treasure. I do not see anyone so no one will discover me and my treasure!

Michael is sailing the ocean when he came upon an island.

Michael: Ah! A deserted island. Is anyone here? Oh look, I’ve discovered a pirate!

find - Active

The thief is hiding behind the tree with the diamond ring.

Thief: I will hide behind this tree. No one is around to find me. I’ll sleep for a little bit.

Lincoln is walking around the park.

Lincoln: I wonder if there is anything behind this tree. Look, I found someone!

Thief: Aak! You found me!

find - Passive

This is a story about Jason and the farmer. The farmer is looking for his tools.

Farmer: hmmm, I wondering if my tools are behind this big doghouse! I can’t see anything so no one will be able to find me behind here.

Jason was walking around the backyard.

Jason: hmmm … . That’s a really big dog house. I’ll just walk around it to get to the other side of the backyard. Oh look, it’s the farmer! I found the farmer.

find - Passive

This is a story about June and Lincoln. June is looking for her diamond ring.

June: I wonder where this ring could be. I’ll look behind this big bench here. It’s hard for people to see me when I’m behind this bench. I’m sure no one will find me here while I’m looking for this ring!

Lincoln is enjoying a lovely day at the park.

Lincoln: What a great walk! I’ll sit down on this bench and rest a bit. Oh, who’s this? Looks like I found my friend, June!

fix - Active

The Grey Robot and the Green Robot are working in the office today. The Grey Robot and the Green Robot accidentally bumped into the cabinets and some of their screws fell onto the floor.

Grey Robot: I only have a pen. Pens won’t fix this problem.

Green Robot: That’s okay, I have a very good screwdriver. I will fix you, Grey Robot!

fix - Passive

The Round Robot and the Square Robot are hanging out in the laboratory one day.

Suddenly, both their arms fell off their body!

Round Robot: I only have a paintbrush. Paintbrushes can’t fix Robots.

Square Robot: Don’t worry, I have a wrench. I can fix your arm!

fix - Passive

The Blue Robot and the Yellow Robot are together in the laboratory mixing chemicals.

There was a big explosion. Looks there are wires everywhere.

Yellow Robot: I only have scissors. Scissors will not fix our wires.

Blue Robot: I have superglue! I will fix you!

forget - Active

Mommy and Audrey are at the mall. Audrey is staring at the toys when Mommy walks off.

Audrey: Mommy! Where are you going? Did you forget me?

Mommy: Oh gosh, it’s true. I’m sorry. I forgot you in the store.

forget - Passive

Lucas and Mary are at the playground. Mary is playing by herself.

Lucas: Hey, I did not forget you! We were playing together last week!

Mary: No, I forgot who I was playing with last week.

forget - Passive

Chase and Chloe are at school. Chase is playing by himself.

Chloe:Hey, I did not forget you! I played at your house last week.

Chase:You did? I forgot who was at my house last week.

frighten - Active

Cole and Aurora are going to a party. Pirates frighten Aurora but Cole loves pirates.

Cole:Your costume is awesome, Aurora. Look, I’m dressed as a pirates. Do I frighten you?

Aurora:Yes, you frighten me! Pirates don’t look very nice.

frighten - Passive

Andrew and Caroline are at a Halloween party. Mummies frighten Caroline but Andrew loves mummies.

Andrew:Look, Caroline, I’m dressed as a mummy. Do I frighten you?

Caroline:Yes, you frighten me! I can barely see you coming for me!

frighten - Passive

Owen and Jackie are at a costume party. Ladybugs frighten Owen but Jackie loves ladybugs.

Jackie: Owen, I love your witch costume! Do you see my ladybug costume? Does it frighten you?

Owen: Yes, Jackie, you frighten me. You know that I don’t like ladybugs!

love - Active

The boy is playing around with his cat. The boy loves it when the cat plays with him.

Boy: oh my goodness, you are so cute, kitty. I love you! Do you love me?

Cat: *hisses* no, I do not love you!

love - Passive

Uncle is babysitting his nephew, Alexander. Uncle and Alexander are talking in the living room.

Uncle: You are the cutest kid I know. I love you very much. Do you love your uncle, Alexander?

Alexander: No! I only love mommy and daddy.

love - Passive

Jake and Isabelle are neighbors. They play with each other every day.

Isabelle: Jake, I don’t love you because I’m jealous of your new clothes.

Jake: But I love you because you’re my only friend, Isabelle!

surprise - Active

Cole and Amelia are best friends. It’s Amelia’s birthday and Cole wants to do something for her.

Cole: Amelia, I want to surprise you! Happy birthday!

Amelia: Yay! I was not expecting this at all. I am so happy!

surprise - Passive

Clara and Owen are at a baseball field. Owen bought something very special for Clara.

Owen: Look, Clara! I want to surprise you. New baseball mitts!

Clara: Thanks, Owen! You really surprised me! I have nothing to surprise you with today.

surprise - Passive

Mommy and Caroline are shopping at the toy store. Mommy bought Caroline something very special.

Mommy: Caroline, I want to surprise you! Look, a new bike!

Caroline: Wow! So cool! But I can’t surprise you, Mommy, you have everything!

spot - Active

The boy is playing around with his pet monkey. When the boy wasn’t looking, the monkey climbed up one of the trees.

Boy: ah, I spot you hiding in the tree, Monkey! You silly monkey.

spot - Passive

Auntie is babysitting her niece, Audrey. Audrey goes behind the curtain.

Audrey: I’m hiding from Auntie behind this big curtain. I can’t see anything but I bet Auntie will not spot me.

Auntie: Oh, there are little shoes poking out from the bottom of the curtain. I spot you, Audrey!

spot - Passive

Uncle is babysitting his niece, Chloe. Chloe goes behind the brooms.

Chloe:I’m hiding from Uncle behind these brooms. I can’t see anything but I bet Uncle will not spot me.

Uncle:Oh, there are pigtails poking out from the brooms. I spot you, Chloe!

wash - Active

Harvey is playing with his dog in the backyard.

Look how dirty they’re getting!

Harvey: You’re so dirty, dog. I need to wash you before we go back into the house.

wash - Passive

Ava and Andrew are in the kitchen. Look! Ava is spilling juice everywhere.

Ava: Aww! There’s juice on my dress. I have nothing to wash this off with!

Andrew: I have a wet towel. Let me wash you, Ava.

wash - Passive

Benji and Anna are in the bathroom. Benji and Anna are dirty from playing outside.

Anna:I can’t find anything to wash you with, Benji.

Benji:I found a sponge! I don’t mind being dirty; I will wash you first, Anna.

Appendix B. Truth-Value Judgment Task Performance by Verb

Figure A1 shows the percentage of correct responses by four-year-olds for individual verbs tested in each lexical semantic profile.

Figure A1. Percentage of correct responses by verb for four-year-olds. (Error bars indicate standard error.)

In general, for each profile, children perform similarly across List A and List B verbs. While there are some notable differences – particularly with Profile 1 (fix vs wash) and Profile 4 verbs (forget vs spot) – these differences may be driven by small sample sizes. In particular, this experiment was a between-subjects design with a sample size of 19 children (11 children for List A and 8 children for List B). So, lower performance for verbs like wash and forget may be driven by only having judgments from 11 and 8 children, respectively.

In addition, the much lower performance of forget (apart from the low sample size of 8 children) could also be due to the experimental test items for that verb – see Appendix A. In particular, the forget test items may have been pragmatically ill-formed. Consider this example test item for the passive use of forget:

Chase and Chloe are at school. Chase is playing by himself.

Chloe: Hey, I did not forget you! I played at your house last week.

Chase: You did? I forgot who was at my house last week.

In this scenario, it could be that children think remember is more pragmatically appropriate as a lead-in than forget (i.e., Chloe would say, “Hey, I remember you! I played at your house last week.”). However, to ensure the materials across all stories for all verbs were uniform, the target verb (here: forget) had to be used in the story leading up to the test sentence. We note that this potentially pragmatically-odd test item was not an issue for adult participants – adults performed near ceiling for forget.

More generally, future studies could more thoroughly investigate children’s performance on the passive of Profile 4 verbs by (i) testing more children, (ii) creating better test items for forget, and (iii) testing more verbs that fall under Profile 4. Additionally, as noted in footnote 9, because test materials could not be constructed to include plausible dissent, we may expect overall comprehension rates to improve across all verbs if tested in a more felicitous setup.

More generally, we want to reiterate (as in the General Discussion) that if four-year-olds do in fact comprehend Profile 4 verbs correctly in the passive, our qualitative results do not change. The four-year-olds would still pattern differently from the three- and five-year-olds, and the lexical semantic hypothesis is still supported as children of different ages seem able to comprehend progressively more lexical semantic profiles in the passive as they get older.

Notes

¹ We note that we chose to focus on long verbal be-passives because short passives (e.g., It was carried) with actional verbs (like carry) are ambiguous between adjectival and verbal passives. This ambiguity has led some prior work, such as Borer & Wexler (1987), to suggest that young children’s early success with actional passives is due to interpreting those passive structures as adjectival rather than verbal be-passives. That is, young children would be using different syntactic structures to successfully interpret these actional passives compared with other passives. By focusing on long verbal be-passives, we can therefore focus on what is generally agreed to be a single underlying passive structure.

² The specific features that Liter et al. (2015) were interested in for their study were EVENTIVE and AGENTIVE. They defined EVENTIVE as a verb being unacceptable in the simple present tense in an “out of the blue” context. Here, we use the opposite definition of EVENTIVE for STATIVE. Also, we have renamed AGENTIVE to VOLITIONAL to avoid terminological overlap with the AGT-PAT feature proposed by Messenger et al. (2012).

³ Stativity can also be defined as a verb being unacceptable in the simple progressive form – for instance, Diana is loving Matthew is less acceptable than Diana is carrying Matthew (Vendler 1957). Here, we follow the definition from Liter et al. (2015).

⁴ Messenger et al. (2012) used the terms “theme-experiencer” and “experiencer-theme” to refer to OBJ-EXP and SUBJ-EXP, respectively.

⁵ Refer to emmanguyenling.github.io for more specific details of the meta-analysis.

⁶ These studies either reported successful performance on these verbs in the active or did not report any difficulty in comprehending the meanings of these verbs. We take this to mean that unsuccessful performance in these studies results from children’s difficulty with these verbs in the verbal passive rather than difficulty with the task or the meanings of the verbs themselves.

⁷ Three children were excluded because they exhibited a yes-/no-bias, and one child was excluded for answering only one of the five active controls correctly.

⁸ Of these 19 children, four children answered four of the five active controls correctly while the remaining 15 children answered all five active controls correctly.

⁹ We note that there was no statistical difference in performance between children who received training and those who did not – see section 4.4 for more discussion .

¹⁰ We note that the test stories did not include plausible dissent in order to keep them uniform to each other and simple enough to allow for multiple test stories within a single session. This may have made all the stories less felicitous, and so potentially caused four-year-olds not to correctly comprehend the passive for a particular verb when they might have comprehended it in a setup with plausible dissent. However, because all stories were alike in this regard, we expect this to have a global effect, potentially lowering correct comprehension rates across all profiles. So, the qualitative pattern we report below likely would remain the same, though perhaps the profile on the borderline (Profile 4) might have moved into being accepted by four-year-olds in a more felicitous setup. We leave this for future work.

¹¹ For discussion on the potential differences in children’s performance between the two verbs in each profile, see Appendix B.

¹² An anonymous reviewer notes that it is possible that some of the included children who answered only four out of the five active controls correctly (4 of the 19) were guessing on the control items. This is because, at an individual level, 80% correct cannot be statistically distinguished from chance performance at the p<0.05 level (using a binomial distribution). But, because this would only account for four children in our sample, we are reasonably confident that the results we find are valid.

¹³ An anonymous reviewer pointed out that some of the Profile 2 verbs such as surprise and frighten have been argued by Hirsch & Hartman (2006) to fall under Belletti & Rizzi’s (1988) “Preoccupare” class of psych verbs. If this is the case, then it is the active sentences of these verbs, rather than the passive, that should be difficult for young children to comprehend. Because children were effectively at ceiling for all active control items, including active sentences with surprise and frighten, we suspect that any predicted difficulty of verbs that fall under this “Preoccupare” class would appear in children younger than the ones we tested (cf. Borga & Snyder (2018) for evidence of French-speaking children’s difficulty with active OBJ-EXP sentences until four years of age).