Publishing Partner: Cambridge University Press CUP Extra Publisher Login

New from Cambridge University Press!


Revitalizing Endangered Languages

Edited by Justyna Olko & Julia Sallabank

Revitalizing Endangered Languages "This guidebook provides ideas and strategies, as well as some background, to help with the effective revitalization of endangered languages. It covers a broad scope of themes including effective planning, benefits, wellbeing, economic aspects, attitudes and ideologies."

New from Wiley!


We Have a New Site!

With the help of your donations we have been making good progress on designing and launching our new website! Check it out at!
***We are still in our beta stages for the new site--if you have any feedback, be sure to let us know at***

Review of  The Comprehension of Relative Clauses by Romance Learners of English: Syntactic and Semantic Influences

Reviewer: Hongying Xu
Book Title: The Comprehension of Relative Clauses by Romance Learners of English: Syntactic and Semantic Influences
Book Author: Marie-Anne C.M. Duffeler
Publisher: Netherlands Graduate School of Linguistics / Landelijke (LOT)
Linguistic Field(s): Applied Linguistics
Language Acquisition
Subject Language(s): English
Language Family(ies): Romance
Issue Number: 29.4520

Discuss this Review
Help on Posting

Marie-Anne Duffeler’s monograph “The Comprehension of Relative Clauses by Romance Learners of English: Syntactic and Semantic Influences” investigates, as the title indicates, the roles that syntax and semantics play in the comprehension of English relative clauses (RCs) by Romance language learners of L2 English: native speakers of French and Italian, to be more exact. Specifically, she investigated the following questions:

a) does the syntactic function of the head nouns in the relative clause affect L2 learners’ comprehension of the RCs; if so, does it conform to the Noun Phrase Accessibility Hierarchy (NPAH) proposed by Keenan and Comrie (1977)?

b) does the position of the preposition in the oblique relative clauses affect learners’ comprehension; i.e., whether the preposition has been moved to the position before the relative pronoun (the pied-piped position) or has stayed internally in the relative clause (stranded position) does L1 transfer play a role?

c) does the semantic meaning of the preposition (locative vs. non-locative) in the oblique RC affect learners’ comprehension?

This monograph is based on her thesis, which investigated the above mentioned research questions. There are altogether six chapters and each chapter will be briefly summarized in the following section. The first chapter introduces the setting of this study. The second chapter summarizes previous studies on RCs and summarizes her research design. The third chapter reports the experiment conducted among French-speaking learners of English on their comprehension of RCs. The fourth chapter reports the experiment done with the group of Italian-speaking learners. The fifth chapter reports the experiment on Italian-speaking L2 English learners’ comprehension of oblique RCs. The last chapter recaps the research questions, findings, and conclusions by putting the three experiments together.

Chapter 1 introduces the motivation of this study based on the importance of syntax acquisition and the gap identified in the current literature on the acquisition of RC by L2 learners. Specifically, the author pointed out the following weaknesses along this line: a) the studies were set up on the basis of research on L1 RCs by employing a diversity of hypotheses, thus they lack focus; b) the skills investigated are mainly procession and production, but not much research exists solely on comprehension; c) most studies focused on the subject-object asymmetry, but not on other types of RCs; d) the methodology employed is very limited: mostly production tasks or grammaticality acceptance tasks; e) there is a lack of consensus on the findings. She then justifies her study, which are intended to expand research along these lines.

Chapter 2 gives a literature review on the research on L1 RCs and L2 RCs and explains the research design of the present study. It starts by summarizing different aspects of RCs, including a definition, different types of RCs, syntactic properties, and a generative analysis of RCs. It goes on to discuss the typological differences regarding the three specific languages involved in the present study: English, French, and Italian. It points out the commonality among these three languages as well as the differences in terms of relativizing strategies, syntactic restrictions, and so on. It then briefly summarizes some theories on second language acquisition and L1 transfer. A large section of this chapter is used to summarize the literature on RC research in L1 and L2. After a historical review of the trends and developments in this field, the author mainly reviews the research on the subject-object asymmetry regarding L1 native adults, children, and L2 learners. Most L1 studies on adults’ processing of RCs as well as on the children’s comprehension reported an advantage of subject RCs over object RCs. A summary of different theoretical accounts to explain this advantage follows, including the NPAH (Keenan & Comrie, 1977), the Perspective Hypothesis (PH, MacWhinney, 1997), the Structure Distance Hypothesis (SDH, O’Grady, 1997), and Frequency of occurrence (Mithell, et al, 1995). The complexity hypothesis proposed by McDaniel et al. (1998) based on Chomsky’s Minimalist Theory was also introduced. The research design section introduces the participants, target RCs and skill investigated, followed by the specific research questions, including the roles played by syntax, semantics and transfer from L1. Hypotheses were made accordingly based on the different accounts respectively: a) there is a gradient of difficulty in comprehension: subject RC has an advantage over object RC, which in turn has an advantage over oblique RC; b) the pie-piped oblique RC has an advantage over stranded oblique RC due to L1 transfer; c) oblique RCs with non-locative preposition have an advantage over those with locative position due to higher cognitive load with spatial sentences.

Chapter 3 presents the experiment design and discusses the results of the comprehension of RCs by French learners of L2 English. Fifty-nine French learners of English at intermediate-low level participated in this experiment. The experiment involved one task: a picture selection task. Participants listened to a request (Show me …) with a RC embedded in it and then circled the head noun of the relative clause from three or four candidates in a picture within 5 seconds. For example, participants would hear: “Show me the tiger that is pushing the horse.” Then they would see a picture in which there are two tigers and one horse: with the horse is the middle. The horse is pushing the tiger in front of it, but is being pushed by the tiger behind it. Participants were expected to circle the tiger that is behind the horse. There are altogether 56 sentences: 16 fillers, 16 subject RCs, 8 object RCs, 8 oblique RCs with pie-piped preposition (OBL-PP RCs), 8 oblique RCs with Stranded Preposition (OBL-PS RCs). Each correct selection got 1 score. The results indicated that there is no significant difference between participants’ performance with the subject RCs and the object RCs. However, participants’ performance with the subject RCs and object RCs were significantly better than OBL RCs. Furthermore, participants did significantly better in comprehending OBL-PS RCs than OBL-PP RCs. Therefore, the first hypothesis was partially supported: the syntactic role of the head noun in the RCs did affect comprehension, but not in the predicated way. The second hypothesis was not supported since opposite results were obtained.

Chapter 4 presents the results of comprehension of RCs by a group of Italian learners of L2 English. Ninety-one students participated and a majority of the participants self-evaluated their proficiency level as intermediate-low or intermediate-high. Two tasks were used with this group: a picture selection task and an elicited imitation task. The picture selection task was very similar to the one described above, except that the number and make-up of the sentences changed. There are altogether 48 sentences: 16 fillers, 8 subject RCs, 8 object RCs, 8 OBL-PP RCs, and 8 OBL-PS RCs. As for the elicited imitation task, participants listened to a sentence once and then were given 40 seconds to write down the sentence. There are 25 sentences: 17 fillers, 4 subject RCs, 4 object RCs. The syllables of each sentence were controlled. The imitation task results were rated by two people. The results from the picture selection indicated that there was no significant difference between participants’ performance with subject RCs and object RCs, nor between object RCs and oblique RCs. However, participants performed significantly better with subject RCs than oblique RCs. Furthermore, there was no significant difference between OBL-PP RCs and OBL-PS RCs. The results from the elicited imitation task did not show any difference between subject RCs and object RCs.

Chapter 5 presents the experiment on the comprehension of oblique RCs by a group of Italian learners of L2 English. It was a 2 * 2 design with two variables: syntax (pie-piped vs. stranded prepositions in RC) and semantics (locative prepositions vs. non-locative prepositions). For example, “over” in “stand over” is a locative preposition, whereas “at” in “look at” is non-locative. Fifty-nine learners participated in this experiment, the majority of which self-evaluated themselves as learners at the intermediate-high level. A picture selection task and an elicited imitation task were used in this experiment with similar procedures, but different sentences. The results from the picture selection task showed that the participants did significantly better with OBL-PS RCs than OBL-PP RCs under both the locative and non-locative preposition settings. They did significantly better with non-locative oblique RCs than locative oblique RCs under both the pip-piped and stranded settings. The results from the elicited imitation task showed a similar finding with the syntactic variable, however, opposite results with the semantic variable: participants did better with the locative ones than the non-locative ones.

Chapter 6 summarizes the whole thesis by restating the research questions, hypotheses, and results and highlighting the conclusions and the implications of this study. To sum up, syntax and semantics both play a role in L2 learners’ comprehension of relative clauses, but not in the exact way predicted by NPAH or other theoretical accounts. The transfer of L1 has not been identified. It is concluded that future research in relative clauses should take more factors into consideration than just the syntactic factor, and should employ more precise methodology to target specific skills in acquiring relative clauses.


This monograph is very well-structured both as a whole book and within each chapter. There are always recapping, referring back, and summaries of content from the previous chapters, which makes it very easy for readers to follow. Most of the terms are consistent throughout the book.

This study has contributed to the research of relative clauses, especially to the research on L2 acquisition of relative clauses in the following ways. First, it presents a comprehensive review of existing research on relative clauses, including linguistic studies, L1 acquisition studies, and L2 acquisition studies. In particular, it has done a thorough investigation and provided a summary of the studies on the subject-object asymmetry among L1 adults, L1 children, and L2 learners. It also provides a good summary of more recent studies on RCs, especially the investigation of factors other than syntax, such as the animiacy of the head nouns and so on. All these will definitely benefit other scholars when they do research along these lines in the future.

Another contribution this study made is the original investigation of frequency of occurrences of the target types of RCs using both a British English corpus and an American English corpus. The findings can be used as reference for future research.

It also made a good analysis on the weakness of previous studies and identified the gap in the literature for future research. For example, previous studies mostly focused on processing or production of RCs, on the subject-object asymmetry. Moreover, the findings from previous studies, L2 RC studies in particular, failed to reach consensus. The research questions investigated in this study are closely connected to the gap identified in the literature review. It was designed to investigate comprehension instead of processing or production of RCs and to investigate both syntactic factors and semantic factors. Moreover, the present study expanded its investigation to the oblique RCs in addition to subject RCs and object RCs.

In designing the picture selection tasks, the animacy and category of the head nouns were controlled to exclude possible confounding factors that had been identified in previous studies. As for the elicited imitation tasks, the length of each sentence in terms of syllables was also controlled for the same reason. All these have contributed to increase the validity of this study.

The author added the elicited imitation task to the two Italian groups in order to eliminate the possible ceiling effect from the picture selection task. Furthermore, the elicited imitation task was creatively used in an innovative way by making participates reproduce the sentences in written form rather than in oral form, which has been the normal practice. This gives scholars another possible option to consider when the imitation task is to be employed.

The data was analyzed with great caution. Friedman’s ANOVA was used since the data from these tasks lacked a normal distribution. Wilcoxon’s signed rank-tests were used with a Borferroni correction for pair-wise comparisons.

However, there are still a few aspects regarding the study goals, experimental designs, and conclusions that need further discussion. First and foremost, this study examined the comprehension of RCs; however, in the literature review part, the previous studies were organized and discussed regarding types of RCs, different factors and variables affecting RC acquisition rather than any specific focus on comprehension studies. The theories based on which the hypotheses regarding the research questions were made are different in nature. For example, the Syntactic Distance Hypothesis and the Minimalism theory are both theories about implicit syntax knowledge, whereas the frequency account is usage-based. It is still unclear if these theories are in a good position to make predictions on L2 learners’ comprehension. In addition, the findings from this study were discussed by comparing to previous studies that explored processing, production, or implicit knowledge of RCs. As has been pointed out by the author, these are different skills and one does not translate into another. Therefore, more references are called for to better understand these findings.

Secondly, this study attempted to investigate whether L1 transfer presents in the comprehension of RCs by Romance learners of English. However, there is not enough literature review or discussion on the theories regarding L1 transfer in L2 acquisition. The Full Transfer Full Access Hypothesis and a usage-based account on L1 transfer were briefly introduced. However, the FTFA Hypothesis only predicts the presence of L1 transfer at the beginning of L2 acquisition. Moreover, the prediction focuses on L2 learners’ implicit knowledge rather than their comprehension skills. In the research question and its hypothesis, it is assumed that since both groups’ L1s only allow the pied-piped position of the preposition in oblique relative clauses, if L1 transfer is present, the pied-piped ones should be easier to understand than the position-stranded ones. This assumption should be taken with caution. On the one hand, even if the predicted advantage is observed, it may be due to a different factor, the frequency factor, for instance. In order to tease apart possible factors, a different group should be included whose native language has an opposition restriction regarding the position of the preposition in the oblique RCs. On the other hand, even if there is absence of the predicted advantage, it still cannot arrive at the conclusion that there is no L1 transfer in L2 acquisition of RCs. What this study has done is take a snapshot of two basically the same groups of learners. With neither comparison of groups with different L1s nor cross-sectional groups within one particular L1, only very limited conclusions can be drawn about L1 transfer.

Last but not least, the results of the experiment that investigated the role of semantics on comprehension of oblique RCs indicated some complication: in the picture selection task, participants’ comprehension was better with non-locative prepositions, whereas in the imitation task, participants did better with the locative prepositions. The author attributed these conflicting findings to the difference between the tasks. The picture selection task is more demanding because participants had to process the locatives and to interpret the pictures at the same time. However, the scores on locatives from the picture selection task were as high as or even higher than those from the imitation task. It seems this explanation does not work well. What really sticks out is the crash down of non-locatives from the imitation task, for which there is no explanation. Therefore, the conclusion that confirms the semantic role in comprehension of oblique RCs needs to be considered with caution.
Hongying Xu is an assistant professor at the University of Wisconsin-La Crosse. She has a Master’s in TESOL and a PhD in Curriculum and Instruction, with special interest in second language acquisition in instructional settings and foreign language pedagogy.