Home Verb Class-Specific Caused-Motion Constructions
Article Publicly Available

Verb Class-Specific Caused-Motion Constructions

  • Xiaorong Xia

    Xiaorong Xia (b. 1973) is a lecturer at the School of Foreign Languages and Cultures, Nanjing Normal University, Nanjing, Jiangsu Province, China. Her research interests include theoretical linguistics, cognitive linguistics and semantics. Publications include “Syntactic variants of caused-motion events” (2014), “The fusion of verbs of contact by impact into caused-motion constructions: A connectionist view of verbs and variants in conceptual frames” (2013), “Break verbs in caused-motion constructions” (2012), “A comparative study on conceptual structure and construction grammar” (2007).

    EMAIL logo
Published/Copyright: November 16, 2017
Become an author with De Gruyter Brill

Abstract

In a constructional approach, the caused-motion construction is productive enough to attract verbs of different types into the construction; however, the distinct senses derived from actual instances indicate it is necessary to posit the caused-motion construction at lower levels because more novel uses present meanings closer to those lower constructions. The present analysis of the corpus data of English motion verbs shows that the senses of manner of caused-motion, manner of causing motion, and accompanied motion arise from their occurrences in the caused-motion construction. From a usage-based perspective, the entrenched use would yield verb class-specific constructions that are productive as well. The lower level of construction, together with the most schematic one, is stored in our memory as part of conceptual representation. The research indicates that creative use of motion verbs in the caused-motion construction is better interpreted with verb class-specific constructions.

1 Introduction

In her study of English caused-motion constructions, Goldberg (1995, 2006) directs her attention to an argument structure construction that pairs the syntactic form NP V NP PP with the semantics X CAUSE Y TO MOVE Z, and identifies several extended senses from the central sense. Van der Leek (1997) criticizes Goldberg for resorting to constructional polysemy in order to avoid lexical polysemy (or verbal polysemy in this case). On the other hand, the extended senses are closely related to verbal semantics and seem to derive from the central sense because of the usage of verbs in the construction. However, within a usage-based model (Langacker 2004a, 2005a, 2005b), linguistic units are abstracted from usage events by reinforcement of recurring commonalities. It should be the case that the same caused-motion construction expresses different senses in actual use, considering different verbal semantics, and then a more schematic construction, for example the argument structure construction described by Goldberg, gradually forms. That is, there might be constructions at lower levels with less schematicity, which are more closely related to verbal semantics. Croft (2003) argues for the productivity of verb class-specific constructions and verb-specific constructions and insists on no necessity for a more schematic construction. On the contrary, Goldberg (1995), Barðdal (2006) believe productivity comes from the most schematic construction.

The present analysis examines corpus data of English caused-motion constructions taken from WebCorp (http://www.webcorp.org.uk). It aims to identify lower levels of caused-motion constructions with different productivity quantified in terms of type frequency and token frequency. The use of verbs of different types or of diverse semantics in caused-motion constructions actually proves the high productivity of the construction; however, the distinct senses derived from actual instances indicate it is necessary to posit caused-motion constructions at lower levels because more novel uses present meanings closer to those lower constructions. The research indicates that verb class-specific constructions can motivate the caused-motion interpretations of different classes of verbs. These constructions from entrenched usage are more responsible for the novel use of non-typical caused-motion verbs.

2 Caused-motion interpretation

Verbs that are not conventionally used in caused-motion constructions may acquire a novel interpretation which is not lexically encoded. Break is a verb of change of state and empty undergoes the conversion from an adjective to a verb indicating change of state. Both are conventionally used in transitive constructions in which the direct object is the affected. Sneeze and laugh are intransitive verbs which do not have a semantic object.

    1. Sam carefully broke the eggs into the bowl. (Goldberg 1995: 21)

    2. The blue boxes are emptied into special trucks. (From Overseas English [Overseas English (2005–2009). ISSN1009-5039/CN 34-1209/G4.])

    3. She sneezed the foam off the cappuccino. (Goldberg 2006: 94)

    4. Frank laughed him out of the room. (Peña 2009: 759)

The license for un-subcategorized arguments in the sentences of (1) has been the focus of research into the relation between verbs and the construction. For lexical semanticists, these additional arguments challenge their hypothesis on the projection from argument structure to syntactic structure. Different syntactic structures would need diverse argument structures of verbs, which is not semantically parsimonious. Constructionists thus suggest the mapping between semantics and syntax is fulfilled via constructions rather than lexical entries or linking rules.

According to Goldberg (1995: 76), caused-motion constructions bear the central sense of X CAUSES Y to MOVE Z together with the form of NP1 VP NP2 PP NP3. This form may correspond to several other senses, being extended ones of the central sense above. The senses (i)–(vi) are exemplified by sentences (2a)–(2f) respectively.

    1. Frank sneezed the tissue off the table.

    2. Mary urged Bill into the house.

    3. Sue let the water out of the bathtub.

    4. Harry locked Joe into the bathroom.

    5. Sam helped him into the car.

    6. Sam accompanied Bob into the room.

    1. X CAUSES Y to MOVE Z

    2. The conditions of satisfaction associated with the action denoted by the predicate entail: X causes Y to move Z.

    3. X ENABLES Y to MOVE Z

    4. X PREVENTS Y from MOVING Z

    5. X HELPS Y to MOVE Z

    6. A further extension (Agent and Theme move along a specified path.)

Goldberg’s definition of some of the caused-motion senses is similar to Johnson’s (1987) classification of the FORCE schema (see Evans & Green 2006: 187–189). The central sense resembles the COMPULSION schema of the FORCE schema, which emerges from the experience of being moved by an external force. The third extended sense is similar to the REMOVAL OF RESTRAINT schema, which captures a situation in which an obstruction to force is removed. The fourth extended sense is related to the BLOCKAGE schema deriving from encounters in which obstacles resist force. But if the caused-motion sense is defined in terms of physical and non-physical force, only the experiences of the COMPULSION schema and the ENABLEMENT schema are involved. (1a) denotes an external physical force exerted by the action encoded in the verb andconforms to the experience of the COMPULSION schema. In contrast, the other sentences in (1) do not involve an external physical force and thus express experiences similar to the ENABLEMENT schema. This schema comes from the sense of potential energy, or lack of it, in relation to the performance of a specific task. The ENABLEMENT schema does not involve an actual force, but a potential force vector.

Though the central sense and the extended senses combine into a category of related senses that are associated with caused-motion constructions, it is not difficult to find that these senses are closely related to certain types of verbs. Except that the central sense is defined with semantic factors such as “manipulative causation and actual movement” (Goldberg 1995: 162), the extended senses are described as related to different classes of verbs. For example, sense (ii) pertains to force-dynamic verbs that encode a communicative action (e.g. order/ask/invite/beckon/urge/send). They can appear in caused-motion constructions only when the conditions associated with actions they denote are satisfied. Sense (iii) derives from instances with force-dynamic verbs encoding removal of a barrier (e.g. allow/let/free/release). Sense (iv), contrary to sense (iii), refers to the imposition of a barrier and relates to verbs like lock/keep/barricade. Sense (v) involves ongoing assistance to move in a certain direction with the use of verbs like help/assist/guide/show/walk. A further extension pertains to a subclass of verbs whose agent is not the cause, enablement, or prevention of motion and whose agent argument and theme argument move along a specified path. Verbs such as accompany/follow/trail/tail fall into this class. Although Goldberg (1995) does not relate the central sense to any class of verbs, we may infer from her examples with push/kick/sneeze/shove that they are force-dynamic verbs denoting the exertion of physical force.

The caused-motion construction is thus polysemous, involving systematically related senses for each class of verbs. That is to say, each extended sense is not only closely related to the central sense but also pertains to certain types of verbs or verbs with certain semantic features. It is inevitable to connect each extension to the semantics of verbs. For example, Kay (1996) defends that the differences in entailment (namely, the extended senses) result from semantic differences among the verb classes, with which constructions will be unified. The meaning of the verb contributes partly to the caused-motion sense. The entailed prevention or possibility of motion is specified in the semantics of a certain class of verbs, but not in the semantic events comprising the caused-motion construction. Thus, the extended sense (iv) can be attributed to the meaning of lock, which inheres ‘prevention of moving out.’

Admittedly, the caused-motion construction is really attractive, considering different classes of verbs occur in it. The verbs that do not bear a caused-motion sense, like sneeze, break, empty, laugh, etc., gain the interpretation in the construction. Motion verbs participate in the construction though they lack cause; the act denoted by verbs of contact by impact does not necessarily involve motion, but a caused-motion reading is feasible when they occur in the construction; even the change of state encoded by break verbs does not block them from the construction.

    1. Learn how to strike the shuttlecock across the net and down into the ground[1]

    2. then you whack the ball onto the green

    3. An ambulance rushed him to Kings County Hospital. (From Overseas English)

    4. Sue swam the baby to the bank. (Ritter & Rosen 1998: 141)

    5. you need to smash the stone from the shelf

    6. differential pressure just to crack the ball off the seat

In the constructional research, verbs occur in constructions which are productive and are thus responsible for providing semantic interpretations. However, as discussed above, the caused-motion senses of these verbs add to new explanations. Alternatively, some researchers take verbal constructions as being responsible for a caused-motion reading. Boas (2003, 2008) advocates a verb, together with its conventional sense, forms a mini-construction. Iwata (2008) posits two verb-specific constructions of spray: [NP spray NP PP] and [NP spray NP (with NP)], bearing the semantics of change of location (as in 4a) and coverage (as in 4b), respectively.

    1. John sprayed paint onto the wall.

    2. John sprayed the wall with paint.

Croft (2003) and Barðdal (2006) contend that constructions exist at different levels of schematicity in the schematicity–lexicality continuum. The lowest level involves the most lexical information, as the name of verb-specific constructions indicate; the level above contains verb subclass-specific constructions; the next level consists of verb class-specific constructions, and then the higher level of event-type constructions, etc. The level above is an abstraction of the level below; the highest level is the most schematic and contains the least lexical information. Researchers differ in the productive level. Barðdal (2006) believes the highest level is productive, that is, the productivity of each construction is derived from the highest construction. Clauser & Croft (1997) take the most entrenched level as the most productive.

The following section observes the corpus data in WebCorp, available from Birmingham City University (http://www.webcorp.org.uk/live/), in order to analyze the novel sense arising from the fusion of motion verbs into caused-motion constructions. We believe that not only verbal sense changes when a verb occurs creatively in the construction; in addition, new interpretation occurs to the constructional sense. We aim to prove that the syntactic form of caused-motion constructions and the derived constructional sense, once entrenched, form verb class or verb subclass-specific constructions which are responsible for interpreting the nonce occurrences of verbs in the same class.

3 Verbs of motion

In Levin (1993: 263–270), verbs of motion consist of verbs of inherently directed motion (e.g. arrive, descend, enter), leave verbs (e.g. abandon, desert, leave), manner of motion verbs (including roll verbs, run verbs), verbs of motion using a vehicle (e.g. skate, fly, drive, ride), waltz verbs (e.g. jig, polka, tango, waltz), chase verbs (e.g. chase, follow), and accompany verbs (e.g. accompany, guide, lead).

A motion event, according to Talmy (2000), is made up of MOTION, FIGURE, GROUND, and PATH, together with optional CAUSE and MANNER. The subclasses of motion verbs specify certain elements in their conceptual structures. Verbs of inherently directed motion specify the direction of motion in the absence of an overt directional complement, while leave verbs only have the direct object denoting the location away from which the motion takes place (Levin 1993: 264). Both denote the action profiling the motion of figure in reference to ground. The two types of verbs discussed above are unacceptable in caused-motion constructions due to the specified GROUND, instead of PATH. The following discussion would not involve the two subclasses.

3.1 Verbs of motion using a vehicle and waltz verbs

Verbs of motion using a vehicle and waltz verbs are found in caused-motion constructions, as seen in (5) (from Levin 1993: 267–269). They express the motion by means of vehicles or different kinds of dance. Some of the former appear with prepositional phrases indicating change of location, and the latter may have the caused-motion sense only if the dancing action they denote involves partners.

    1. He skated Penny around the rink.

    2. He rowed Penny across the lake.

    3. He waltzed her across the floor. (If dance involves a partner.)

The motion expressed by the verbs depends on vehicles or the manners of dance, and involves accompanied motion as well. The skating action and the rowing action need the motion of the skater and the rower together with Penny in order to transport her to the destination; similarly, the two dancers move in a waltzing manner. The construal of passengers and dancers into two sides turns the motion event into a caused-motion event, and thus the verbs fuse into the construction.

Verbs of motion using a vehicle cover those that are vehicle names and those that are not. The latter have more occurrences of the caused-motion construction, except cruise and oar. The former can be further divided into subclasses using different vehicles: boat (canoe, ferry, gondola), cycle (bicycle, moped, motorcycle), sledge (bobsled, skateboard, toboggan), cab (jeep, taxi, trolley), and flying craft (helicopter, jet, parachute). The difference does not have any influence on the verbs that occur in caused-motion constructions; only a few (parachute, rocket, boat, bus, taxi, tram, trolley, sledge) occasionally express a caused-motion sense.

    1. They had the parachuters parachute the balls into the centre of the field/We sledge the swan across the shingle./it was suggested that they canoe the vehicles across the river/ask if I would also be willing to cycle the torch across the Forth Road Bridge

    2. These smaller craft will in turn taxi the persons to the larger ships overhead./It is a pretty good pose and he did rocket the ball out of the park

3.2 Chase verbs and accompany verbs

Chase verbs and accompany verbs can all occur in caused-motion constructions. They indicate more as to the relation between the agent and the theme. The two types of verbs likewise denote a different case of motion, namely, two sides of participants are in motion in the action. We chase somebody in motion, and then we are in motion, too. In the case of companionship, the two participants are in motion nearly at the same time.

Compare the following examples. Ann’s and Sam’s motion can only be said to be on the same path with the theme’s motion.

    1. Sam accompanied Bob into the room.

    2. Ann chased the squirrel out of her house.

In a strict sense, chase verbs and accompany verbs in the syntactic form of NP V NP PP do not express a typical caused-motion sense, given that the two participants are both movers, neither causing the motion of the other.

3.3 Manner of motion verbs

Roll verbs and run verbs express manners or means of motion. Levin (1993) points out that roll verbs have the inanimate theme while most run verbs describe the manner in which the animate theme moves. The verbs encoding the motion of the inanimate theme are more often used to express a caused-motion sense, as indicated by the corpus data. All roll verbs are found with prepositions profiling different portions on the path of motion and most have quite a few instances. Nevertheless, verbs of rolling (drift, drop, float, glide, move, roll, slide, swing) have higher token frequency in the construction than coil verbs. We think it is the designated path encoded in the lexical senses of coil verbs that hinders more appearances. Coil verbs denote motion around an axis, as indicated by most instances with around.

    1. …you wind the strings on the post

    2. Then he proceeds to twirl the upper on the lower stick

    3. Then coil the cord into the bucket for easy carrying.

Only some run verbs can occur in caused-motion constructions with one or two instances, however. Run verbs indicate different manners of motion, such as moving at a high speed (charge, dart, dash), moving in files (file, march, parade), moving by leaps and bounds (jump, lope, vault), moving in a stealthy manner (sidle, skulk, sneak), moving along quickly and lightly (flit, fly, skitter), moving in a leisurely manner (sashay, tramp, hike), etc. The verbs expressing quick motion have the most occurrences in caused-motion constructions among all the run verbs. Among the 25 verbs, 9 occur with more than three different types of prepositions, and only 5 have no occurrences. In contrast, just a few of the verbs of other motion manners occur in the construction occasionally.

    1. Oh, couldst thou know how gladly dart the fish across the sea

    2. Shepherds parade the sheep through the city every year in order to exercise

    3. When Phoebe tries to sneak the dog out of the apartment

    4. The men leap the horses across the field

As shown by the corpus data, verbs encoding manner or means (vehicle) of motion are more likely to occur in caused-motion constructions than the others. The specification of the manner or means apparently more easily activates the causation in the motion event. However, we observe that the caused-motion interpretation is different from the prototypical one. At least three extensions can be derived from the instances with motion verbs. The next section will go more deeply into them.

4 Manner of caused motion

Verbs of rolling express the manner of the theme’s motion when they are used in caused-motion constructions. Rolling and bouncing actions involve the manner of the ball’s motion in the following examples.

    1. I rolled the ball into the hole.

    2. I bounced the ball across the room.

The motion event encoded in the semantics of verbs of rolling implies the action of a force exerter who initiates the motion of the theme. That is to say, the ball will not move in a specific manner unless external force applies on it.

Rolling is the manner of an object changing its location. Usually, the entity rolls because of an external force applied on it.[2] Verbs of rolling usually indicate the motion of the inanimate. It is human action that can possibly lead to the motion. Their fusion into the caused-motion construction needs an additional causer who initiates the motion. The speaker’s or the conceptualizer’s construal of the rolling action has to establish the relation between the entity in the motion event and another participant in the event of causing the motion. That is, in order to construe the motion action into a caused-motion event, windowing of attention has to be extended to an initiator who causes the motion. In comparison, we enter or walk into the room because our muscles coordinate for this purpose. No physical force is externally applied on us as far as the walking action is concerned.

Likewise, the use of coil verbs in caused-motion constructions involves the construal of an extended participant as initiator of the motion. As indicated by the sentences in (11a), the ball absorbing the force moves to the designated goal and is intended to stay there.

    1. to apply horizontal force to bounce the ball toward the target/bounce the ball to the child, call out a word

    2. over to the left to revolve the turret to the left---and so on./and you can revolve the back to the position you want; vertical, horizontal or/he could coil the neck toward the torso/Working from each end simultaneously, coil the dough toward the center, forming an S shape.

Though a causative relation is established between participants in the motion event, no specific information as to the manner of causation is profiled. All the roll verbs have the sense of ‘cause to move’ in the dictionary, but their caused-motion instances give a reading of X CAUSE Y TO MOVE Z in a SPECIFIC MANNER. Different from typical caused-motion instances, which profile the manner of causing the motion, they describe the manner of the caused motion. The prepositional phrase in caused-motion constructions only implies the direction or location of movement, whereas the manner of motion is usually unmarked. The verbs like throw, carry, break, empty, squeeze, etc. usually denote the manner of causing the motion, but that of motion of the moved is not specified. If any, it must be a default manner of motion: falling under gravity. But in (10, 11), the theme’s manner of motion is profiled. The sense, undoubtedly, derives from that of roll verbs.

Some run verbs have the same interpretation in caused-motion constructions. In general, run verbs refer to human manner of motion, animal manner of motion, and ball manner of motion. Some verbs, like run, refer to the manners of motion of both humans and animals. We would like to further divide the class into two subclasses, jump verbs and walk verbs. The division does not rely on the different manners of motion; rather, the two subclasses have different senses in caused-motion constructions. Similar to roll verbs, jump verbs in caused-motion constructions refer to the manner of motion of the theme but not that of the agent. The following sentences imply that something done by the agents causes the motion of the animals. The verbs can just be interpreted as ‘cause to trot/run/jump.’

    1. She trotted the horse into the stable.

    2. The scientist ran the rats through the maze.

    3. Sylvia jumped the horse over the fence.

We know it is the moved entities, namely, the animals, that undergo the change of location in a specific manner. If the action denoted by these verbs is put in a panoramic view, the external cause would appear. In the case of the rats running through the maze, the speaker’s attention extends to the scientist, who is responsible for the rats’ motion. In any case, the scientist is the initiator who directly stirs their motion and there is a relation between them. However, no attention is paid to his action, namely, the manner of causation in which the rats are moved.

5 Manner of causing motion

Walk verbs, on the other hand, acquire a distinct interpretation in caused-motion constructions. Goldberg (1995: 162) thinks “She walked him to the car” expresses the extended sense (v) of the caused-motion construction, indicating that the woman’s special walking gait assists the man’s motion to the car. Similarly, we can interpret (13a, 13b) as ‘a person’s walking causes the letter/the ladder to move’ by conceiving the agents’ manner of motion helps with the movement of the inanimate.

    1. John walked the letter to the dean’s office. (From Overseas English)

    2. Sharon walked the ladder across the room. (Langacker 2004b: 291)

However, (14) has a different implication.

  1. He walked the dog along the street.

When we walk a dog, we know sometimes it is the dog that drags us to walk quickly or we have to follow it at a fast pace. That is to say, his walking in (14) might not be the assistance for the dog’s motion. Ritter & Rosen (1998: 157) propose not treating walk in (14) as derived from the basic walk by a productive process of syntactic insertion of an internal argument, but rather a distinct form with a different semantic representation with the sense similar to exercise.

In fact, walk has already got the dictionary interpretation of ‘cause to move.’ The examples in (15) are listed in the Oxford English–Chinese Dictionary (1988: 1291, Peking: The Commercial Press) for this sense. This sense makes (15a), the passive form of walk, accepted. But it is not the case with (13a, 13b). The reason might be the inanimacy of the theme, which cannot move in a walking manner.

    1. The horse should be walked for a while after a race.

    2. He walked his horse up the hill.

    3. He put his arm around me and walked me away/off.

    1. ?The letter was walked to the dean’s office.

    2. ?The ladder was walked across the room.

In addition, walk is ambiguous in Goldberg’s sentence and (14), but not in (13a, 13b). The former can be either interpreted as she just walked, accompanying the man as he walked to the car, or that her hands supported the man and thus caused him to walk to the car. It is the same with the man and the dog in (14). The ambiguity is also clear in (15b), in which the rider might hold the reins and walk together with the horse or ride on the horse and walk it by means of spurs, or the rider might give it a whip and therefore the horse moves up the hill. As for (13a, 13b), since the inanimate theme cannot walk, they just have the sense (v). That is, walk denotes the causer’s manner of motion in (13a, 13b) while in Goldberg’s sentence and (14) it may refer either to the causer’s manner of motion or the theme’s manner of motion. But if walking refers to the causer’s manner of motion in (15b), it is clear that just his walking cannot cause the horse to move. Rather, there must also be something special accompanying his walking, for example his whipping or pulling the reins while walking, that causes the horse’s motion.

The sense of assistance is more clearly seen in (17a): without the walking of Fred, who supported the drunk, the latter’s change of location wouldn’t have taken place. It is the direct object’s inability to move steadily that intensifies the assistance of Fred’s walking which is crucial to the change of position.

    1. Fred walked the semi-conscious drunk across the street. (Langacker 2004b: 291)

    2. I walked her to the bus stop. (From the scripts of American TV series Friends)

However, (17b), which is quite similar to Goldberg’s sentence, reveals little assistance. One of the heroes in the TV series recalls a fond memory of his ex-wife and himself on an occasion when he went with her to the bus stop on foot. The sentence does not indicate that he gave her any assistance in walking, nor does he do anything to cause her to walk. He just walks together with her to the bus stop. Therefore, the assistance reading does not cover all the examples listed above. (13) and (17a) express a sense that X moving in a specific manner causes Y to move Z, namely X CAUSE (in a MANNER of MOTION) Y TO MOVE Z. This entailment is similar to the prototypical caused-motion sense, except that the causer’s manner of motion is specified. Inevitably, the specification is related to the semantics of the verbs. We will come back to the sense of the other instances discussed here.

6 Accompanied motion

Accompanied motion is an extension from the caused-motion sense, in that the relation between participants is different from that in the case of causative motion. In the latter, one participant acts as the cause of another’s motion on a path, the former being initiator and the latter being mover; in the case of accompanied motion, the two participants are both movers on the same path. In a word, causation is not so apparent as co-motion.

Furthermore, the sense of accompanied motion is specified by motion verbs in different ways. Accompany verbs just indicate the companionship between movers; chase verbs imply the motion of one participant being followed by that of another, namely, the order of movers; verbs of motion using a vehicle and waltz verbs express accompanied motion on the same vehicle or in the same dancing.

  1. Actor1+ MOVE accompany/chase/skate/waltz + Actor2 + Location path

The different subclasses of verbs manifest the different means of motion, either by accompanying, chasing, using vehicles, or dancing. The second Actor is optional in the skating and waltzing event; the absence of Actor 2 would see the argument realize into intransitive structures, like They skated across the street/They waltzed across the room. However, the elaboration on the actors, or to put it more exactly, on the role of the participants, would make it possible to construe the participants in a causative relation. In the skating event, the skater might be the driver and/or the passenger; if the role of the skater is profiled, the event can be construed as the control of one skater causing the motion of passengers to the goal. Both driver and passengers change their positions. Similarly, in the dancing event which involves two dancers, one side (usually the man) gives signals to the other side by force from his hands and the latter would move as he expects. The skating driver and the leading dancer can be construed as causers, while the passengers and the partner can be construed as themes in motion. The construal would activate a causative relation between actors. (19) describes the elaboration of participants in an accompanied-motion event in the case of moving on a certain vehicle or by means of certain dancing. The participants are elaborated into Controller and Passenger (or Load) in the former, and into Leader and Partner in the latter.

    1. Controller + MOVE vehicle + Passenger/Load + Location path

    2. Leader + MOVE dance + Partner + Location path

7 Verbal constructions

Participating in caused-motion constructions, motion verbs do alter the caused-motion sense. In the usage-based model (Langacker 2005a, 2005b), lexical items represent abstractions from actual usage just as constructions do. A new sense might arise from repeated occurrences of the verb in a construction. Verbs and constructions are characterized by schemas with different abstraction or specificity. If constructions exist at different levels with graded schematicity, it is reasonable to attribute extended caused-motion senses to verbal constructions. The entrenched usage of roll verbs, chase verbs, and accompany verbs yields senses pertaining to the verb classes; that is, verb subclass-specific constructions are formed pairing the senses and the syntactic form Vmotion NP PP NP. The more creative occurrences of run verbs do not contribute more senses; their caused-motion readings fall onto the two senses. The senses are inherited from the caused-motion construction which has the most schematicity, and at the same time are closely related to the semantics of verbs. (20) shows the syntax and the semantics of the roll verb construction.

  1. NP + ROLL VERB + NP + PP

    CAUSER + (CAUSE) + MOVER + MOVE + PATH

The manner of CAUSE is gapped, but a causative relation exists between CAUSER and MOVER. The syntactic form is similar to that of the caused-motion construction; they differ in semantics in that the roll verb construction has the sense X CAUSE Y TO MOVE (in a SPECIFIC MANNER) Z.

On the other hand, the accompanied-motion construction is derived from the great number of instances with chase verbs and accompany verbs. Verbs of motion using a vehicle, waltz verbs, chase verbs, and accompany verbs occur in caused-motion constructions when participants in the motion event are elaborated into two sides in causative relation. The construction is described as follows.

  1. NP + ACCOMPANY VERB + NP + PP

    MOVER1 + ACCOMPANY(in a specific manner of motion) + MOVER2 + PATH

The accompany verb construction, though similar to the caused-motion construction in syntax, presents a sense that seems different from the prototypical caused-motion sense, in that no force exertion exists in the causative relation.

Compared to the two senses discussed above, the sense derived from the occurrence of walk verbs in (13) and (17a) is closer to the central caused-motion sense, namely, the causer’s motion causes the subsequent motion of the caused. However, we think the sense expressed by the other instances of walk verbs is more than assistance, as we have mentioned in section 5. There is something common in all the instances; the motion event is achieved by both movers, that is, they move together. No matter whether we walk the lover to the bus stop, or we walk the letter to the dean, we move in company with the latter. Then an accompanied motion is also available in the occurrences of walk in the caused-motion construction. In addition, the usage of other walk verbs can be interpreted by the accompanied-motion construction.

    1. An ambulance rushed him to Kings County Hospital. (From Overseas English)

    2. Sue swam the baby to the bank. (Ritter & Rosen 1998: 141)

Moreover, the two verb subclass-specific constructions can interpret the nonce instances of other run verbs.

  1. Accompanied motion

    1. Then take the shuttle or amble the trails through the forest to the restored Chellberg Farm/Some of the horses stomp the trim off the foot/When Phoebe tries to sneak the dog out of the apartment/saw emergency workers race the baby out of the backyard/as he appeared to shuffle the ball off the line with his arm Manner of caused motion

    2. The men will swim oxen across the river, one team at a time./To bowl the ball toward the plastic bottles/when they canter the horses across the field/4 points on the hill test requires that you coast the bike through the key section at a range of different entry speeds/I had to creep the car off the freeway/After numerous attempts, I inch the car through the metal gate posts/Hurtle the debris into the valley below

The sentences in (23a) show the motion of the agent, together with an entity, in a specific manner; those in (23b) have the sense of motion of the theme, which results from the agent’s action. Describing a man who shuffled the ball off the line (see the underlined sentences above), we aim to indicate something as to his manner of motion, rather than that of the ball. His specific manner of motion causes the motion of the ball, but both participants move. In contrast, when we hurtle the debris into the valley below, we do not fall quickly into the valley; rather, it is the debris that will move at a high speed.

As far as the manners of motion are concerned, motion at a low speed (inch), motion by crawling, rolling, and sliding are often construed as the manner of caused motion; motion in heavy steps (stomp) and motion in a stealthy way (sneak) are apt to be construed as the manner of accompanied motion. The two latter manners of motion might be more related to human beings, whose action is more likely to trigger the motion of other entities. Similarly, walk, which dominantly expresses human manner of motion, shows the agent’s manner of accompanied motion in the corpus data.

  1. Walk the bike across the crosswalk./Strap yourself in or walk the ball across the beautiful slopes of memorial park/Walk the bike toward the incline until it begins to roll on its own/After the conference, have someone walk the parent to the front office exit if possible./the minute I walk the freak out of the office

Verbs expressing motion at a high speed occur in caused-motion constructions more often than the other verbs, and this motion can be construed either as the manner of causing motion or as that of caused motion. Gallop and hurry in (25) indicate either the manner of the theme’s movement or that of the agent’s. The verbs have both the sense of ‘cause to move at a high speed’ and that of ‘move quickly to bring about the motion of another entity.’

The manner of the actors’ fast motion might more easily spark accompanied motion.

    1. An actor appearing as the groom had to gallop the horse across the stage/gallop the mail to the next station

    2. Hurry the bread into the hot oven/It was Jordan’s decision, in fact, to hurry the prisoner through the backdoor into an awaiting squad

8 Conclusion

When participating in caused-motion constructions, motion verbs do alter the caused-motion sense. The verbs which have a higher occurrence represent the alteration. The first altered sense is X (in a SPECIFIC MANNER OF MOTION) ACCOMPANY Y TO MOVE Z, characterized by the caused-motion instances of chase verbs and accompany verbs; the second one is X CAUSE Y TO MOVE Z in a SPECIFIC MANNER OF MOTION, which is clear in the instances with roll verbs.

In the constructional research, verbs occur in constructions which are productive and are thus responsible for providing semantic interpretations. The use of motion verbs in caused-motion constructions cannot be attributed exclusively to the construction which is supposed to provide interpretations for verbs. The semantics of the verbs also contributes to some degree. The repeated use of this class of verbs in caused-motion constructions yields a verb subclass-specific construction. As indicated in the usage-based theory, these lower levels of constructions are stored as well as the most schematic one, and are likewise available in the process of production and perception.

About the author

Xiaorong Xia

Xiaorong Xia (b. 1973) is a lecturer at the School of Foreign Languages and Cultures, Nanjing Normal University, Nanjing, Jiangsu Province, China. Her research interests include theoretical linguistics, cognitive linguistics and semantics. Publications include “Syntactic variants of caused-motion events” (2014), “The fusion of verbs of contact by impact into caused-motion constructions: A connectionist view of verbs and variants in conceptual frames” (2013), “Break verbs in caused-motion constructions” (2012), “A comparative study on conceptual structure and construction grammar” (2007).

References

Barðdal, Jóhanna. 2006. Construction-specific properties of syntactic subjects in Icelandic and German. Cognitive Linguistics 17(1). 39–106.10.1515/COG.2006.002Search in Google Scholar

Boas, Hans C. 2003. A lexical-constructional account of the locative alternation. In Leslie Carmichael, Scott C.-H. Huang, & Vida Samiian (eds.), Proceedings of the 2001 Western Conference in Linguistics, 27–42. Fresno, CA: California State University Publications.Search in Google Scholar

Boas, Hans C. 2008. Resolving form-meaning discrepancies in Construction Grammar. In Jaakko Leino (ed.), Constructional reorganization, 11–36. Amsterdam/Philadelphia: John Benjamins Publishing Company.10.1075/cal.5.02boaSearch in Google Scholar

Clausner, Timothy C., & William Croft. 1997. Productivity & schematicity in metaphors. Cognitive Science 21(3). 47–282.10.1207/s15516709cog2103_1Search in Google Scholar

Croft, William. 2003. Lexical rules vs. constructions: a false dichotomy. In Hubert Cuyckens, Thomas Berg, René Dirven & Klaus-Uwe Panther (eds.), Motivation in languages: Studies in honor of Günter Radden, 49–68. Amsterdam/Philadelphia: John Benjamins Publishing Company.10.1075/cilt.243.07croSearch in Google Scholar

Evans, Vyvyan & Melanie Green. 2006. Cognitive linguistics. Edinburgh: Edinburgh University Press.Search in Google Scholar

Goldberg, Adele E. 1995. Constructions: A Construction grammar approach to argument structure. Chicago/London: The University of Chicago Press.Search in Google Scholar

Goldberg, Adele E. 2006. Constructions at work. Oxford/New York: Oxford University Press.Search in Google Scholar

Iwata, Seizi. 2008. Locative alternation: A lexical-constructional approach. Amsterdam/Philadelphia: John Benjamins Publishing Company.10.1075/cal.6Search in Google Scholar

Kay, Paul. 1996. Argument structure: Causative ABC constructions. http://www.icsi.berkeley.edu/~kay/bcg/5/lec05.html (accessed 18 December 2006).Search in Google Scholar

Langacker, Ronald W. 2004a [1984]. Foundations of cognitive grammar (Vol. I). Peking: Peking University Press.Search in Google Scholar

Langacker, Ronald W. 2004b [1984]. Foundations of cognitive grammar (Vol. II). Peking: Peking University Press.Search in Google Scholar

Langacker, Ronald W. 2005a. Construction grammars: Cognitive, radical, and less so. In Francisco J. Ruiz de Mendoza Ibáñez & M. Sandra Peña Cervel (eds.), Cognitive Linguistics, 101–159. Berlin: Mouton de Gruyter.10.1515/9783110197716.1.101Search in Google Scholar

Langacker, Ronald W. 2005b. Integration, grammaticization, and constructional meaning. In Mirjam Fied & Hans C. Boas (eds.), Grammatical constructions, 157–189. Amsterdam/Philadelphia: John Benjamins Publishing Company.10.1075/cal.4.11lanSearch in Google Scholar

Levin, Beth. 1993. English verb classes and alternations. Chicago and London: The University of Chicago Press.Search in Google Scholar

Peña, Ma Sandra. 2009. Constraints on subsumption in the caused-motion construction. Language Sciences 31. 740–765.10.1016/j.langsci.2009.05.003Search in Google Scholar

Ritter, Elizabeth & Sara T. Rosen. 1998. Delimiting events in syntax. In Miriam Butt & Wilhelm Gender (eds.), The projection of arguments: Lexical and compositional factors, 135–164. Stanford: CSLI Publications.Search in Google Scholar

Talmy, Leonard. 2000. Toward a cognitive semantics (Vol. I). Cambridge, MA: MIT Press.10.7551/mitpress/6847.001.0001Search in Google Scholar

Van der Leek, Frederike. 1997. Caused-motion and the “bottom-up” role of grammar. In Ad Foolen & Frederike van der Leek (eds.), Constructions in cognitive linguistics, 301–331. Amsterdam/Philadelphia: John Benjamins Publishing Company.10.1075/cilt.178.17leeSearch in Google Scholar

Published Online: 2017-11-16
Published in Print: 2017-08-28

© 2017 Walter de Gruyter GmbH, Berlin/Boston

Articles in the same Issue

  1. Frontmatter
  2. Part One: Perspectives of Meaning-making
  3. Meaning-Centrism in Roland Barthes’ Structuralism
  4. Part One: Perspectives of Meaning-making
  5. Peirce’s Semiotics in a Commercial Context
  6. Part One: Perspectives of Meaning-making
  7. Once Upon a Time… Fairy Tales and Other Stories
  8. Part Two: Languages and Meaning
  9. Bridging the Unbridgeable
  10. Part Two: Languages and Meaning
  11. Verb Class-Specific Caused-Motion Constructions
  12. Part Two: Languages and Meaning
  13. Properties of Mandarin Reflexives
Downloaded on 23.11.2025 from https://www.degruyterbrill.com/document/doi/10.1515/css-2017-0016/html?lang=en
Scroll to top button