LINGUIST List 25.3039

Thu Jul 24 2014

Review: Cognitive Science; Historical Ling; Linguistic Theories; Semantics; Typology: Goschler & Stefanowitsch (2013)

Editor for this issue: Malgorzata Cavar <>

Date: 31-Mar-2014
From: Konrad Szczesniak <>
Subject: Variation and Change in the Encoding of Motion Events
E-mail this message to a friend

Discuss this message

Book announced at

EDITOR: Juliana Goschler
EDITOR: Anatol Stefanowitsch
TITLE: Variation and Change in the Encoding of Motion Events
SERIES TITLE: Human Cognitive Processing 41
PUBLISHER: John Benjamins
YEAR: 2013

REVIEWER: Konrad Szczesniak, University of Silesia


Edited by Juliana Goschler and Anatol Stefanowitsch, “Variation and Change in the Encoding of Motion Events” is a volume with a clear agenda. One common goal of this collection of studies is to demonstrate the continuum nature of the binary distinction between satellite-framed and verb-framed modes of expression of motion proposed by Talmy (2000a) (2000b). The authors in the volume argue that it is too simple and untenable to divide languages into those that express path in the verb (enter the room) and those that express manner in the verb and path outside it (hop into the room). As Kopecka notes, Talmy’s division, “although clearly fruitful, nevertheless proved too simplistic to account for the typological complexity of individual languages.” (p. 164) Thus, the point being made is that in its traditional form, Talmy’s division may be another facile dichotomy; for maximum accuracy, it should be carefully qualified and shown to exhibit considerable variation. This objective is pursued coherently in two ways, namely through analysis of examples of variation within the two types of languages, and through tracing diachronic change as languages cross the division between the types over time. This approach is certainly reasonable and compelling: the cases of diachronic change offered here are only possible if the distinction is a continuum; no seismic changes between satellite-framed and verb-framed style would be possible overnight as “one-fell-swoop” transformations.

The volume is a fascinating collection of studies focusing on how Talmy’s typology of motion events plays out in languages when looked at in detail, based on large numbers of actual uses of motion event phrases. Berthele justifies this approach by pointing out that “[t]he typological status of a particular language is to be determined empirically, based on corpora, and not via introspection or via genealogical inheritance.” (p. 58) The contributions in the volume provide ample concrete data to evaluate and revise our thinking about how languages capture motion. Each chapter offers a study of uses reflecting the way motion is encoded in languages chosen by the authors; additionally the authors include a distillation of their views and findings accumulated over years of research on the expression of motion in language.


Filipocić shows that in Serbo-Croatian, a satellite-framed language, the conflation of manner and directionality is constrained by aspect and morphosyntax. Unlike in English, where seemingly any manner verb can be used with a path satellite, Serbo-Croatian resists uses of verbs such as skakutati (‘skip’) in patterns like “She was skipping into the house”, because morphological blocking makes it impossible to adjust the verb’s aspect to the boundary-crossing scenario found in such motion descriptions. In other words, this and other Slavic languages are not fully satellite-framed languages. Hijazo-Gascón and Ibarretxe-Antuñano make similar qualifications, albeit about languages found on the other extreme of the continuum. Here too, the main finding is that textbook examples of verb-framed languages--Spanish, French and Italian--are not fully verb-framed languages after all. They have “Manner verb+adverb structures”. The authors conclude that the three Romance languages “are verb-framed languages, but they show intratypological variation with regard to the semantic component of Path” (p.50). Similarly, Berthele looks at the production by speakers of French and a number of contact varieties of German and Romansch, and uses a wealth of statistical data to demonstrate that languages do not differ from each other in a simple binary fashion. Instead, satellite-framed German exhibits signs of verb-framed behavior, while speakers of the Romance verb-framed counterparts French and Romansch allow elements typical of satellite-framed languages such as relatively high numbers of manner verbs with path descriptions. Wälchli and Sölling take a more panoramic look at motion expression across a wide range of languages (although they do not describe each one in detail in their chapter, they have compared 117 languages). They also conclude that there are few universal properties. Compounding the impression of cross-linguistic variation is the observation that “[n]o conclusive evidence for underlying global semantic features such as path was found.” (p.109) They go on to venture that many features in motion typology are “cross-linguistic comparative concepts designed by typologists and not intrinsic in language structure.” (p.110) The next two chapters look at how learners of satellite-framed languages express motion in these target languages. Goschler studies Turkish learners of German, while Jensen and Cadierno look at Turkish and German learners of Danish. Perhaps the most important finding in both studies is that when the mother tongue and the target language are typologically different, learners do not use patterns of motion expression typical of the target language. That is, both studies converge on the conclusion that speakers of Turkish (verb-framed) do not use manner verbs in German or Danish (both satellite-framed) in proportions comparable the production by German or Danish speakers.

Then in the second part of the volume, the authors of four chapters look at how languages changed their modes of motion expression over centuries. Kopecka traces diachronic changes in the development from Old to Modern French, and shows how French lost its focus on Path, thus becoming more of a verb-framed language. On the other hand, Nikitina shows that Greek went in the opposite direction in its evolution from Archaic to Classical Greek. Her point of interest is the use of motion verbs with path phrases in the accusative and dative case, the former being directional and the latter locative. She shows that Greek gradually relied more on the truly goal-encoding accusative phrases, thus becoming a more consistently satellite-framed language. Huber studies changes in Middle English as it incorporated French path-conflating verbs (e.g. enter, ascend) into its motion construction. She provides data to show that these verbs were initially used in the same pattern as other verbs, with prepositional path phrases, but were eventually allotted separate constructions. Stefanowitch’s approach is to treat motion event patterns as a group of largely independent constructions. Instead of viewing a language as being either satellite-framed or verb-framed, one can study motion constructions available in that language. Under this analysis, for example, Romance languages are found to use some German-type patterns where manner verbs appear with prepositional phrases expressing directional motion (e.g. Rose courait au bas de l'escalier ‘Rose ran to the foot of the stairs’). In a way, this should not be surprising, because as Stefanowitsch notes, “most languages have both path- and manner-verbs in their lexicon and allow both the path-in-verb and the path-outside-verb pattern in actual usage.” (p. 226)


The overall picture is one of differences of degree, not kind. A language is not either/or, but tending toward the satellite- or verb-framed end of the continuum. To substantiate this view, the authors identify many factors that complicate the binary typology of motion event expression. For example, Filipović explains limited satellite-framed expression found in Serbo-Croatian in terms of morphosyntactic constraints responsible for disallowing many manner verbs in path-satellite patterns. The constraints she identifies apply not only to Serbo-Croatian, but to similar cases in Polish and Czech, and probably other Slavic languages too. Berthele identifies a correlation between speech community size and the number of manner and path verbs used by the speakers. When their speech communities are small, satellite-framed languages do not necessarily have to feature a large variety of verbs. That is because in small tight-knit communities, more common ground knowledge is shared, “more information can be taken for granted and less explicit forms of utterances are licensed”. As a result, less lexical precision is necessary and fewer types of verbs are used. Also, the incidence of verbs depends on the speakers’ “verbal intelligence” (p.67). This is certainly a fair observation especially in the case of manner verbs, given that there are incomparably more manner verbs than path verbs; the choice of the former requires more effort, and indeed more creativity and eloquence, than the choice between options like ‘go’, ‘leave’, and ‘enter’.

However, while the amount of data and the thoroughness of analyses offered here are truly impressive, it seems that part of the impression of variation is an artifact of the criteria used to classify a language having satellite-framed properties. Some contributors in the volume conclude that a language exhibits signs of satellite-framed behavior based on examples of sentences where manner is conflated with any path phrases. For example, Berthele observes that speakers of Romance (verb-framed) languages in his sample (French and three varieties of Romansch) use manner verbs with path phrases and provides examples like

(1) il saute sur la ruche

‘he jumps onto the beehive’ (example 5, p. 62)

Similarly, Stefanowitsch gives example (2) below

(2) Rose courait au bas de l'escalier. (example 3b, p. 226)

‘Rose ran to the foot of the stairs.’

Huber gives an example from Middle English (3)

(3)Hors þat evir trottid.. It were hard to make hym aftir to ambill well.

‘A horse that ever trotted - it would be hard to make it amble well afterwards.’ (example 1a, p. 206)

While the status of English as a satellite-framed language is secure enough and does not need to be defended, the choice of examples is questionable. A mere use of manner verbs with path phrases is a rather relaxed standard, indeed so undemanding that probably any language can meet it. Under this criterion, even a classic verb-framed language like Spanish allows manner verbs directionally:

(4) La botella flotó en la dirección del mar.

The bottle floated in the direction of the sea.

A true litmus test is whether a language allows manner verbs to appear with phrases that express the resultative element of boundary crossing (Aske, 1989). When this criterion is applied, much of the purported variation disappears -- verb-framed languages cease to exhibit satellite-framed properties. While French can indeed afford uses like (2) above, its satellite-framed potential ends when it comes to expressing equivalents of sentences like ‘Rose ran into the room’ (translated literally as ‘Rose courait dans la pièce’ can only mean ‘Rose ran inside the room’, not ‘into the room’).

Also rather dubious are some conclusions drawn from reports of variation at the opposite end of the continuum, where satellite-framed languages can be observed to show signs of verb-framed behavior. Some authors note that satellite-framed English also uses the verb-framed system (through verbs like ‘exit’, ‘pass’, or ‘ascend’), and they point out that, strictly under Talmy’s typology, this seems to be against the very nature of English. For example, Stefanowitsch reports that in the literature, these verbs cause some consternation as an “oddity in an otherwise pure path-outside-verb language” (p. 229). I believe this is a misunderstanding, a result of treating the two systems on an equal footing. The path-in-verb system involving use of verbs like ‘exit’ or ‘enter’ is probably available to all languages, satellite-framed and verb-framed alike. It is simple standard equipment found not only in classic verb-framed languages, but also in English and other satellite-framed languages. On the other hand, the path-outside-verb system found in satellite-framed languages is a special feature, a more complex way of encoding motion. One could propose an implicational universal along the lines of “If a language is capable of encoding motion satellite-framed style, it also has the option of expressing motion by means of the verb-framed system.” If this approach is accurate, any variation on the satellite-framed side of the division is no longer variation; the choice between the two modes of motion expression is a fairly unsurprising option, like the freedom to occasionally resort to barter trade. Conversely, a verb-framed language cannot and does not show true signs of satellite-framed behavior.

It is not my intention to dismiss the variation advocated in the contributions. It is clear enough that, as the authors demonstrate, it is a hallmark of motion event encoding. The examples of variation reported in each contribution--as well as their discussion--are intriguing, pleasantly stimulating, and indeed genuinely enlightening, and will certainly be welcomed by cognitive linguists, typologists and generally all those interested in the linguistic expression of motion. However, perhaps too much is being made of the presence of variation across Talmy’s divide. To my mind, the variation does not rule out a clear division; it does not justify writing off Talmy’s binary typology as simplistic and replacing it with a continuum view.


Aske, J., 1989. Path predicates in English and Spanish: A closer look.. In: Proceedings of the Fifteenth Annual Meeting of the Berkeley Linguistics Society. Berkeley: BLS, pp. 1-14.

Talmy, L., 2000a. Toward A Cognitive Semantics. Volume I: Concept Structuring Systems. Cambridge, MA: MIT Press.

Talmy, L., 2000b. Toward A Cognitive Semantics. Volume II: Typology and Process in Concept Structuring. Cambridge, MA: MIT Press.


I am currently involved in work on grammatical constructions within the Construction Grammar framework. I am particularly interested in questions of meaning in schematic grammatical constructions. I attempt to reconcile new cognitive approaches with traditional views on questions such as the division into closed- and open-class forms, the lexicon and syntax, and the kinds of meanings that language forms are capable of conveying.

Page Updated: 24-Jul-2014