Grammatical complements as well as lexical collocations create vast semantic implications for “give” and “get”. Take for example the following sentence, “Steve got the question from his teacher” and “Steve got to question his teacher”. The grammatical choice and placement of the preposition complements alone can alter the semantics of the entire utterance. How the learner is able to understand and even produce these kinds of sentences is a matter for testing and the study of the internal lexicon.
Miller and Fellbaum (1992) display the notion: although there are three times as many nouns as are verbs, verbs are arguably the most important lexical category of a language (cited in Aitchison, 1994: p. 110). A vast majority of Japanese learners of English at the tertiary level, having had six years of mombusho (Education Ministry) language training seem to have difficulty when reproducing polyvalent or transitive verbs (written or spoken) such as “give” and “get”, especially when they require a direct and/or indirect object, and more so if a preposition phrase is required.
Craik and Lockhart (1972) state that “better learning will take place when a deeper level of semantic processing is required, because the words are encoded with elaboration” (cited in Schmitt and McCarthy, 1997: p. 242). In order to foster a deeper processing, the students may not be getting valuable exposure to both syntactic and semantic range.
Brown (2000) states that “the principal task of linguistics is to investigate and describe the ways in which words can be combined and manipulated to convey meanings.” (cited in Aitchison, 1994: p. 26). The various aspects of the syntactic and semantic manipulation of a language can be overwhelming, especially for learners at the “false beginner” stage as most of the subjects of this study can be regarded. The linguistic commonalities and differences of English and Japanese, and more specifically the grammatical relationship of “give” and “get” will be investigated in order to provide further insight into how language manipulation can be assisted or hindered.
Moving on to psycholinguistics and the mental lexicon, the basic idea of psycholinguistics is to test and analyze learner responses in order to explore the properties of the mental grammar (Jackendoff, 1994: p. 49). Responses can give researchers an inside look at how learners are formulating and processing language in their mind. There are many ways to test a learner’s language ability and common tests include cloze tests and word associations, to name a few. The common errors as well as common patterns of correctness may give a better understanding of the problems, their sources, and a way to possibly overcome these problems. The three tests used in this study included a translation from L1 to L2, target sentence production in the L2 elicited through pictures, and a cloze test.
The translation from L1 to L2 was chosen as a way to see free production by the students. The translation tasks targeted the verbs in past, passive, and present tenses. All of the questions targeted the verbs with direct and/or indirect objects, in other words transitive and ditransitive states. The picture elicitation questions asked the students to produce sentences once again but with the sentence head provided. Each sentence head along with the accompanying picture aimed to target polyvalent verb responses of “give” and “get”. All pictures showed the image of a ball being transferred between two people, with a highlighted source (agent) and a destination (goal). The last set of questions was designed to test students on their ability to distinguish between the two verbs. Various grammatical tenses were used in combination with familiar lexical terms. This section of tests was presented last in order to prevent exposure to correct syntax prior to the production stages.
The responses from the first two questions provided a general and functional representation of how the students produce the L2, how they use their lexical and syntactic knowledge to create sentences. The more complex nature of production, as well as possible student stress, may not however, fully reflect how the students understand the language. The cloze test was designed to provide further evidence in a more controlled situation. The cloze test gives us a narrower view into how the students understand the use of the verbs in question. The results of the three tests combined should provide a detailed window into the language processing and possible semantic or syntactic areas of concern.
Regarding verbs, there is little substance to language transfer, it is the core of the sentence and is central to the idea of dependency grammar (Lyons, 1977). Dependency grammar, and similarly case grammar, represents the theory that a sentence is dependent upon the verb and its accompanying semantic relationships with noun phrases (Richards, Platt, & Platt, 1992: pp. 46-47).
The syntactic roles of “give” and “get” and their varying semantic roles depending on accompanying complements are the focus of this paper. The number of noun phrases accompanying the verb reflects its valency. “Get” and “give” can be considered polyvalent because more than one noun phrase will usually (in their prototypical meaning) accompany the verb. “Tony gave the girl a bike” is an example of “give” polyvalency because there are three noun phrases (Tony/the girl/a bike) accompanying the verb.
Similar to valency, verbs can also be classified according to their transitiveness or intransitiveness, by verb complement, be it a gerund, infinitive, noun phrase (object), adverbial, adjective, or prepositional complement (Browne, McCray, & Srinivasan, 2000). The complements with respect to this study are demonstrated through their prototypical syntactic and semantic behavior in the transfer of property. In the prototypical sense, “give” and “get” are transitive in nature. While an intransitive verb has no direct object, a transitive verb has a direct object, and a ditransitive verb has direct and indirect object as part of a sentence. One of the difficulties my students have with “give” and “get” may be due to the complex nature of these verbs and their required sentence complements, often involving the prepositions needed to mark the movement of the object or theme of the sentence.
The Japanese English learner has many hurdles to overcome and one of the most basic reasons may be the verb and complement placement. The subject-verb-object order of English is quite different from Japanese grammar that follows a subject-object-verb order. The aim of this study is to find out if a more comprehensive pedagogical practice that encourages students to see and experiment with the fuller ranges of verbs and their complements could help students to understand and produce these verbs more enthusiastically, confidently, and competently.
2. Literature Review
Some interesting recent studies have been done in relation to the effects of proficiency on syntactic priming in the language production of Japanese English as a foreign language (EFL) learner (Hamada & Yokokawa, 2019). They discovered that the alignment at every linguistic level has a significant role in relation to realizing communication goals. Their results showed that upper-level learners were much more capable of being more significantly primed than lower-level learners. Spoken primes appear to activate cognitive links between nodes and lemmas which help upper-level learners in particular to develop their syntactic representations.
This study has adopted the assumption that the prototypical meaning of “give” and “get” represent the transfer of property. In Gronemeyer’s (1999) work, she takes “get” plus noun phrase to be the most basic construction, dominating use of the verb in 39 percent of the examples in the Brown Corpus (Gronemeyer, 1999: pp. 11-12). Gronemeyer (1999) goes on to say that the “prototype” for “get” is equal to ingressive + “have”, which include resultive and inchoative components. Ingressive + “have” basically represents coming to have possession. Resultive or a change of property is the main focus for this paper, while inchoative, as suggested by Kimball (1973), Baron (1977), and Johansson & Oksefjell (1996), involves the emergence of a new state or come to have/be (cited in Gronemeyer, 1999: p. 12). An example of the inchoative would be “get cold” or “get upset”.
Additionally, the phrasal verbs and idioms associated with “give” and “get” that do not carry the prototypical semantic meaning or syntactic structure and will not be addressed in detail. A more ambiguous line can be seen however in phrases such as “give a break” or “get a break” where there is a transfer of something, though it is not a prototypical physical object. However similar and relevant to the overall understanding of how “give” and “get” function linguistically, the main purpose of this paper is to find out how the prototypical semantic and syntactic uses are being produced by the subjects.
With respect to the transfer of property or coming to possess, “give” and “get” are the prototypical semantic representation, and one of the earliest forms of expressing the transfer of property. Since “give “and “get” are relative opposites, they can be said to be quite commonly associated with each other in native speakers’ mental lexicon. This type of opposition can also be called a converse such as in “over/under” (Coulthard, Knowles, Moon, & Deignan, 2000: p. 23). Problems may arise, however, due to the fact that “give” and “get” do not function similarly when used in the passive tense because “get” simply cannot be “gotten”. Verbs are prototypically considered words of action and converse verbs or actions generally take place at the same time or moment.
This temporal unity is also illustrated in “sell” and “buy”. The verbs describe the simultaneity of transfer of ownership. The same event is seen from two different perspectives, that of buyer and seller (Hatch & Brown, 1995: p. 66). With respect to “give” and “get”, if one gives something then simultaneously something is being received. If this event occurs with the same characters involved, then we can demonstrate that when “Chris gives the ball to Ikumi” then simultaneously this semantically infers that “Ikumi gets the ball from Chris.”
The basic in assumption as expressed by Fillmore (1977) is that the semantic difference is not in the transfer that takes place but rather the perspective that is offered by the sentence construction (cited in Vogel, 1998: p. 114). In Table 1, we can illustrate how an identical event can be altered by syntactic perspective.
In the first two examples we see “give” and “get” in basic ditransitive forms (valency of three), having a direct object and indirect object in addition to the noun phrase of the agent. The direct object in the first two examples is a noun phrase and the indirect object is a prepositional phrase with noun phrase, dually related to the verb phrase with Chris and Ikumi as the subject/agent of the sentence. The third example is the same as the first, except the verb phrase is the opposite, and there is no indirect object (giver) and there is no prepositional clue, indicating that the acting verb is passive.
In the fourth example the sentence head is the ball, and this shows semantic and syntactic overlap since the prepositional phrase (by Chris) has to have semantic meaning with the verb phrase and subject (give Ikumi). Chris cannot “give by Ikumi” and therefore, “was given,” would be employed by the native language user. Similarly, though “the ball” is the head of the sentence, the ball cannot “give a person” due to its inanimate nature, not to mention the idea of a person being “given”.
When it comes to valency and transitiveness, they are like grammatical cousins, both define how a verb is complemented by direct and/or indirect objects. A sentence where the verb has a valency of one has no object, such as “she blushes”. This case is quite rare with “give” and “get” because they more commonly require an object, such as in “I gave/I got a flower.” These have a valency of two and are transitive. The sentence, “Steve gave a flower to his mother” has a valency of three and is ditransitive because it requires an indirect object as well as a direct object. Valency and transitiveness are options for grammatical expression of an event. “Mary got a flower” and “Mary was given a flower” show the two verbs transmitting the message of a transfer of property with different syntactic perspectives only.
Allerton (1982: pp. 32-33) states that problems with the valency of verbs can also be seen by semantic plausibility in addition to post-verb sequences. The question of plausibility can be demonstrated by the example “Mike gave Jessica a
Table 1. Verb and its complements (English).
coat of paint” and “Jessica gave the door a bike.” Though syntax is correct, semantically the situation is impossible. The problem students may face is that choosing one object does not restrict the locution on the choice of the indirect object (Allerton, 1982: p. 33). If students are producing sentences such as “Chris gave Ikumi to the ball,” though the syntax is correct, they are not showing real semantic understanding of the subject/object/indirect object relationship.
We will now investigate how the post-verb sequences, direct and indirect objects are cemented to the verb and the rest of the sentence.
Now it is important to consider verb complements. As discussed earlier, the terms intransitive, transitive, and ditransitive refer to the number and type of complements in a verb phrase (Browne, McCray, & Srinivasan, 2000: p. 35). There are many complements to the verb and they are generally known to be noun phrases, preposition phrases, adjectival phrases, adverbial phrases, finite and non-finite phrases, past participle phrases, infinitive phrases, and wh-complements (Browne, McCray, & Srinivasan, 2000). This study will delve further into these complements if and when they arise later in the findings. For general purposes and due to the structure of the test and target responses, we will look at the target complements involving noun and preposition phrases.
1) Ken gave Barbie a ring—VP-V*NP*NP
2) Ken gave a ring to Barbie—VP-V*NP*PP(to)
Number one is an example of a verb phrase with noun phrase plus noun phrase, therefore two noun phrase complements. Number two is a verb phrase with two phrase complements but one is a noun phrase and the other is a preposition phrase. If we take the passive perspective, we have the same pattern as number two but the sentence head is reversed. These verb complements act as glue to the sentence, semantically and syntactically, the complements helping give meaning and definition to the phrase.
Another verb, such as make, has so many meanings that it is difficult to see how it might be defined independent of the words that surround it (Hatch & Brown, 1995: p. 76). Piotrowski (1990) demonstrates how the verb’s surroundings are linked by independent meanings, lexical and grammar-bound collocates, phrasal verbs and idioms (cited in Hatch & Brown, 1995: pp. 76-77). The verbs “give” and “get” act similarly, with distinct phrasal verbs and idioms that are generally learned as chunks.
Noun phrases are generally a lexical matter but the prepositions that demonstrate semantic notions such as direction, place, and origin can cause the second language learner problems. Prepositions such as from, into, to, out show the difficulty in making sharp divides, and to in particular because when semantically representing “towards” it acts as a content word (Aitchison, 1994: p. 107). This should make it easier to understand for the Japanese learner but possibly the confusion may therefore lie in the position of “to” within the sentence.
The learners may understand well what the preposition means or does, and if so the problem of production would be syntactic expression. How does the English syntax and semantic fields correspond to what the language learner is dealing with in their first language? We will now look at the relative equivalent of the Japanese language structure with respect to “give” and “get” and some verb complements. The roles of these preposition phrases along with the verbs and other complements must be considered as they appear in the native learner language.
Now we will examine the Japanese equivalent and language transfer. “Ageru, okuru, sashiageru, ataeru, moshiageru” are commonly used forms of “give”, and “ukeru,uketoru, eru, kureru, morau” are common expressions of “get”. The translation questions used the basic (colloquial) form of “morau” for “get,” “ageru” and “okuru” for “give.” It must be noted that for the closest equivalent to the English to Japanese example below does not correlate with well with “natural sounding” Japanese. The passive example “Mike was given a bicycle,” requires the verb “ataeru,” and this verb semantically implies that the receiver is of lower class or is a little child. Japanese colloquial interpretation of “give” is “ageru” and “get” is “morau” or in more polite situations with “kureru.” (Ishii, M., personal communication Jan, 2005). The use of okuru in the passive is generally used in written discourse and is somewhat awkward in colloquial discourse. Okuru or “give/present,” however, was deliberately chosen because of its phonetic L1 ambiguity with okuru meaning “send.”
In Japanese, the transfer of property verb use can be complicated by the often used combination with another verb, with the adjoining verb being placed ahead and changed to imperative form. If someone bought you a present and you received it then “katte kureta” or “it was bought for me”, creating a passive form in the English translation. But generally “(kare kara) furowa wo kureta” means simply “I got a flower from him.” The agent or source “Kare kara” or “from him” can often be omitted while still implying that it was from someone. This of course would often be dependent on what had been implied or said in the previous exchange. This may all have implications as Japanese learners cope with the direct or indirect object as well as the accompanying prepositions complements. Below is a relative equivalent of the English examples from section 3.3.
We can see that if the heads (subjects) of these sentences and the direct object remain in relatively the same position but the indirect object and the verb positions are reversed. When comparing the example thematic roles of the sentences above we can see how they generally differ from English to Japanese. From top to bottom they are as follows:
In Japanese (see Table 2), the particle “ni” can act similar to the English prepositions in/at/on/to (Bleiler, 1963: p. 26). The preposition or postposition “kara” acts almost identically to the English “from” as a source’s origin (Takebayashi, 1996: p. 192), but also can act as a point of origin in time or space, provenance, movement from a place, the first item in a series, or cause or reason (Takebayashi, 1996: p. 165). “Ha” or sometimes shown as “wa” is the surface marker of the sentence topic but sometimes can act as an exclamation marker (Taylor, 1979: p. 179).
The examples above show “ha” as the topic marker. “Wo” or sometimes seen Romanized as “o” is the particle indicating the previous word as the direct object of the verb (Bleiler, 1963: p. 22). While in Japanese there are no articles, the preposition and particles are extremely important as indicators of the direction or source of the transference of the object/thing. Since the Japanese learner has comparable prepositional meaning and syntactic function between the L1 and L2, problems may originate in the preposition position since the Japanese preposition acts as a postposition. Table 3 clearly highlights the different sentence roles of Japanese and English and how they compare.
With respect to the passive form it must be noted that Japanese has a true passive form and is formed by modifying the verb stem depending on the negative stem (Bleiler, 1963: p. 61). The passive form of “贈る/okuru” would then require dropping the last vowel and adding “-られる/–rareru.” Possessing the passive form provides the Japanese learner with a comparable syntactic framework, how the learner processes this from the L1 to the L2 is the problem. Sometimes however, the passive form in Japanese can be avoided by promoting a theme to subject (usually a noun phrase) and therefore no verb morphology is necessary.
Table 2. Verb and its complements (Japanese).
Table 3. Comparison of English and Japanese sentence roles.
3. Materials and Methods
The test was administered to 96 students with fourteen questions separated into three sections, completed within ten minutes. Questions 1 to 3 were treated as separate tests and were collected prior to the start of each section. Question 3, the cloze test, was administered last because if the cloze test had been administered first it could have provided a framework for the students and led them in their more complex demonstrations required by the translation and sentence making.
The test questions were as follows:
Question 1a—target response “Steve got a present from Tony” (goal-theme-agent).
Question 1b—target response “Tony gave a present to Steve” (agent-theme-goal).
Question 1c—target response “This present was given to Steve by Tony” (theme-goal-agent).
Question 1d—target response “Children get presents at/on Christmas (day)” (goal-theme)
Question 2a—target response “Chris gave the ball to Ikumi” (agent-theme-goal)
Question 2b—target response “Chris got the ball from Ikumi” (goal-theme-agent)
Question 2c—target response “The ball is given to Chris by Ikumi” (theme-goal-agent)
Question 2d—target response “Ikumi gets/got the ball from Chris” (goal-theme-agent)
Question 3.1 “It is a custom in Canada to a gift to your family at Christmas.”
Question 3.2 “Tony’s boss him a new sofa and television for his birthday.”
Question 3.3 “Tony a new CD from his younger sister at Christmas.”
Question 3.4 “Sue was a bag for her 16th birthday.”
Question 3.5 “At the end of the wedding, Tony and his wife wine glasses and a box of chocolates to all of the guests.”
Question 3.6 “Steve his vegetables from the supermarket.”
4. Results & Discussion
A full exploration of the results and error patterns can be found in the appendix. Generally speaking, however, it appears as if the students are having an extremely hard time reproducing the correct usage of both “give” and “get” in the L2. More specifically, Table 4 is a chart indicating the kind of error noted in each question as well as the number of times that kind of error was found within the 96 subject responses. It should also be noted that more than one error could be tabulated within one subject response. For questions 3.1 to 3.6 there are no complex error patterns to be tabulated. Due to the nature of the cloze testing structure, only a single lexical response was required. Non-valid responses represent a non-target verb choice that was plausibly correct.
Overall, the questions that required the students to create a sentence with “get” (1a, 2b, 2d) caused the most difficulty retaining the correct agent (approximately 128 times). Interestingly, “give” (1b, 2a) targeted questions caused substantially fewer agent errors (7 times). Obviously, the students were having much greater difficulty with the verb “get” and the agent-theme-goal structure. With respect to the passive targeted sentences of 1c and 2c, the theme (the present, the ball) as the head of the sentence caused errors (25 times) in which the theme became the agent. Question 1c and 2c responses showed a syntactic pattern where 10% had at least one preposition missing altogether, leaving the semantics of the sentence lacking sometimes a goal, theme, or agent. The passive target question 3.4 caused the greatest number of students to choose the opposite verb in over one third (34 times) of the responses. Clearly, the passive expression was of most concern in all of the responses but now we will look in detail at the salient types of errors that were encountered in each set of questions through thematic and syntactic relations.
Questions 2a - 2d elicit responses using a picture showing Chris and Ikumi interacting with a ball. *Due to the nature of the picture elicitation, almost any verb tense is acceptable though the sentence head for each question guides them to use the target responses.
Table 4. Error count.
**unintelligible response due to word order.
Question 3.1 caused little problems for the students who chose the correct verb (66). Of the 66 correct verbs only 4 students made an inflection error. Questions 3.2 and 3.3 had the greatest number of verb-tense errors. This may have been due to the fact that there was no obvious past tense marker in either one. Question 3.4, most students chose “give” (50) yet were correct in their choice of passive only 22 times. Question 3.5 the student responses were equally erroneous with respect to tense and verb choice. In question 3.6 a majority of the students (32) forgot to inflect the verb in the third person when using the present tense.
Rutherford (1987: chs.1, 2) states that learner exposure be not simply a presentation of solutions but should be in a manner that takes advantage of the learner’s cognitive capacity. Hulstin and Laufer (2001: p. 539) reiterate this point, saying that long term memory retention of vocabulary was highest among comprehension tasks, lower in the reading plus fill-in, and lowest in the reading only groups. A highly grammar-based textbook may not be allowing students to forge the best links in the understanding and production of the L2.
With relative converse verbs like “give” and “get”, care must be taken if they are taught together. Schmitt and McCarthy (1997: p. 232) go on to say that “if students are taught left and right together, they will probably remember the directions, but perhaps not which is which”. Nation (1990: p. 47) also suggests that 25 percent of words will typically be cross-associated (cited in Schmitt and McCarthy, 1997: p. 232) when taught together, and in the case of opposite verbs, cause the learner to choose one verb when actually wanting to choose the other verb. Early learners should be given not necessarily a wider range of syntactic verb structures, such as in variety of verb complements, but possibly a broader scope of the prototypical syntax of these common verbs. Showing how they act with prepositions is a very important aspect in being able to make sense of and produce competent sentences and utterances. That combined with a syllabus challenging the students with L2 production may help decrease the prevalent nature of learners fossilizing bad syntactic habits.
In hindsight, the test design could have been improved in order to help define error patterns a little more clearly. Question 3.2 maybe needed a time marker such as “last week” to help the students identify that the past tense was required. The same could be said for 3.5, “after the wedding” may have been a more helpful phrase to indicate the past tense. But the tense error was only one facet of how the students were unable to recognize and use the correct syntax. Otherwise, the responses provided a great deal of information about how these students seemed to be processing these polyvalent verbs “give” and “get”.
At the tertiary level, there is a lot more opportunity to expose Japanese learners to English as a second language along wider pedagogical lines. With this opportunity in hand, I had intended to re-test the same students at a later time after half of the subjects had received consciousness-raising tasks while the other half had received direct teacher to student answer instruction. The students enjoyed the c-r tasks exponentially more than the traditional answer instruction. As part of the c-r, the students were faced with the challenge of finding the correct answers together and then presenting them to me, giving them a social and interactive experience as they found the answers together. If the c-r groups had incorrect answer(s), noun inflections and article errors aside, they had to return to their group and find out what and where the errors were. All students in the c-r groups were responsible for writing and participating. Undoubtedly a follow-up test would have greatly enhanced this study and possibly proven greater improvements in the use of the target verbs among the c-r groups.
The rigid teaching approach in Japan has restricted most students from using the second language freely and creatively, thus the learners appear to be at times bewildered when confronted with tasks of language production. Additionally, with respect to this study, the students had difficulty in even identifying which verb to choose when faced with the option of “give” or “get.” A more comprehensive pedagogical practice that encourages students to see and experiment with the fuller ranges of verbs and their complements could help students to understand and produce these verbs more enthusiastically, confidently, and competently.
 Jackendoff, R. (1994). Word Meanings and What It Takes to Learn Them: Reflections on the Piaget-Chomsky Debate. In W. F. Overton, & D. S. Palermo (Eds.), The Jean Piaget Symposium Series. The Nature and Ontogenesis of Meaning (pp. 129-144). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. (Reprinted in Modified form from R. Jackendoff, “Languages of the Mind,” MIT Press, 1992)
 Johansson, S., & Oksefjell, S. (1996). Towards a Unified Account of the Syntax and Semantics of GET. In J. Thomas, & M. Short (Eds.), Using Corpora for Language Research (pp. 57-55). London: Longman.
 Piotrowski, T. (1990). The Meaning-Text Model of Language and Practical Lexicography. In J. Tomaszczyk, & B. Lewandowska-Tomaszczyk (Eds.), Meaning and Lexicography (pp. 277-286). Amsterdam and Philadelphia, PA: John Benjamins.