30.3270, Review: Morphology; Semantics; Syntax: Barðdal, Pat-El, Carey (2018)

[The LINGUIST List]

LINGUIST List: Vol-30-3270. Thu Aug 29 2019. ISSN: 1069 - 4875.

Subject: 30.3270, Review: Morphology; Semantics; Syntax: Barðdal, Pat-El, Carey (2018)

Moderator: Malgorzata E. Cavar (linguist at linguistlist.org)
Student Moderator: Jeremy Coburn
Managing Editor: Becca Morris
Team: Helen Aristar-Dry, Everett Green, Sarah Robinson, Peace Han, Nils Hjortnaes, Yiwen Zhang, Julian Dietrich
Jobs: jobs at linguistlist.org | Conferences: callconf at linguistlist.org | Pubs: pubs at linguistlist.org

Homepage: http://linguistlist.org

Please support the LL editors and operation with a donation at:
           https://funddrive.linguistlist.org/donate/

Editor for this issue: Jeremy Coburn <jecoburn at linguistlist.org>
================================================================


Date: Thu, 29 Aug 2019 11:16:07
From: Sune Gregersen [s.h.g.rygard at uva.nl]
Subject: Non-Canonically Case-Marked Subjects

 
Discuss this message:
http://linguistlist.org/pubs/reviews/get-review.cfm?subid=36513697


Book announced at http://linguistlist.org/issues/30/30-453.html

EDITOR: Jóhanna  Barðdal
EDITOR: Na'ama  Pat-El
EDITOR: Stephen Mark  Carey
TITLE: Non-Canonically Case-Marked Subjects
SUBTITLE: The Reykjavík-Eyjafjallajökull papers
SERIES TITLE: Studies in Language Companion Series 200
PUBLISHER: John Benjamins
YEAR: 2018

REVIEWER: Sune Gregersen, University of Amsterdam

SUMMARY

The ten papers collected in ‘Non-Canonically Case-Marked Subjects: The
Reykjavík-Eyjafjallajökull Papers’ are the product of a conference held at the
University of Iceland in 2012. The contributions concern various aspects of
non-canonical subject constructions in a number of different languages, both
ancient and modern. The language families represented are Semitic,
Nakh-Daghestanian, and several branches of Indo-European. The volume consists
of an introduction, three main parts, and an ‘Afterword’. The introductory
Chapter 1 by Jóhanna Barðdal (1–19) summarises the contents of the book and
briefly outlines some of the open questions and unresolved debates in the
study of non-canonical subjects. The three main parts contain papers devoted
to areal and genealogical investigations (Chapters 2 and 3), synchronic
investigations of a more theoretical nature (Chapters 4 and 5), and four
investigations of historical languages (Chapters 6–9). The two chapters in the
final section (Chapters 10 and 11) make a number of theoretical points about
subjecthood and subject tests.

In the first contribution (Chapter 2), Victor A. Friedman and Brian D. Joseph
provide a survey of ‘Non-nominative and depersonalized subjects in the
Balkans: Areality vs. genealogy’ (23–53). The languages investigated include
Albanian, Romani, Greek, and a number of Romance and Slavic languages of the
Balkans. As the title indicates, the authors have cast the net widely and
included not just non-nominative subjects, but a range of ‘impersonal’
constructions, such as weather verbs and other natural phenomena, impersonal
passives, and ‘narrative imperatives’ where the subject is a canonical
nominative argument but an imperative is used instead of the usual verb form.
A number of areal patterns are identified involving these constructions. In a
core area in the central Balkans (present-day Albania, Kosovo, North
Macedonia, and adjacent areas) all of the constructions are found, whereas
some of them are absent in more peripheral languages. For instance, the
narrative imperative is not found in Greek, and both Greek and Romanian lack
the ‘internal disposition’ impersonal found in other languages (such as
Macedonian “mi se jade” [me.DAT REFL eat.PRS.3SG] ‘I feel like eating’, p.
40).

Chapter 3, by Bernard Comrie, Diana Porker, and Zaira Khalilova, investigates
‘Affective constructions in Tsezic languages’ (55–82). The five Tsezic
languages – Bezhta, Hinuq, Hunzib, Khwarshi, and Tsez – belong to the
Nakh-Daghestanian (or East Caucasian) language family and are spoken in the
highlands of southwestern Daghestan. They all have a number of ‘affective’
transitive predicates, such as verbs meaning ‘like’, ‘see’, ‘know’, and
‘forget’, where the most prominent argument is in the dative or one of the
local cases rather than the canonical ergative. The authors survey the
morphosyntactic behaviour of such affective predicates across the five
languages, pointing out differences and similarities between these and
canonical ergative-absolutive predicates. They conclude that the dative (or
local case) arguments in affective constructions share more properties with
canonical A and S arguments than with P arguments, such as their behaviour
with respect to valency change, relative clauses, and reciprocals, and can
thus reasonably be considered syntactic subjects.

The two chapters in the ‘Synchronic’ part are both devoted to the analysis of
non-canonical subject-like arguments within Role and Reference Grammar (RRG).
In Chapter 4, Patrick Farrell and Beatriz Willgohs propose ‘A macrorole
approach to dative subjects’ (83–113), specifically experiencer arguments in
Spanish marked with the dative preposition “a”. The authors note that while
such arguments pattern with canonical nominative arguments in many ways, there
are also important differences, e.g. in passive and control constructions.
They propose a revised version of RRG which recognises three rather than two
macroroles: In addition to the established RRG macroroles Actor and Undergoer
(the prototypical agent-like and patient-like arguments), Farrell & Willgohs
propose a third macrorole, which they term ‘Receptor’. This covers both the
recipient argument of ditransitive predicates, the causee in causative
constructions, and dative experiencer arguments marked with “a”. In this way
the notion of ‘non-canonical subject’ in Spanish can be done away with, and
all dative arguments are recognised as belonging to the same semantic
macro-category. Interestingly, the other contribution on RRG, ‘Dative case and
oblique subjects’ by Robert D. Van Valin, Jr. (Chapter 5; 115–131), reaches a
very different conclusion on the same topic. Based on a survey of the
functions of dative cases in a number of languages, Van Valin argues that the
dative is better analysed as a ‘default’ case without any semantic content of
its own, which is assigned when an argument does not belong to either of the
two macroroles Actor and Undergoer. In other words, the various functions of
datives across languages do not reflect a common semantic core (such as
Farrell & Willgohs’s ‘Receptor’ macrorole), but are related only by virtue of
not being Actors and Undergoers.

‘Word order as a subject test in Old Icelandic’ (135–154) by Jóhannes G.
Jónsson is the sixth chapter and the first contribution to the ‘Diachronic’
part. The paper investigates the word order properties of subjects and objects
in Old Icelandic prose. While it is generally agreed that oblique subjects in
Modern Icelandic are indeed subjects (cf. e.g. Thráinsson 2007, ch. 4; Barddal
& Eythórsson, this volume), there is less agreement about the Old Icelandic
situation. Jónsson shows that while the word order of Old Icelandic is
flexible in some respects, it is possible to use it as a diagnostic for
subjecthood when the arguments of a transitive predicate occur postverbally.
With a few principled exceptions, the word order in such cases is always VSO.
Jónsson concludes that dative-nominative predicates such as the verb “líka”
‘like’ have dative subjects, since the postverbal order is dative before
nominative: “Eigi líkaði honum það vel” [NEG liked he.DAT that.NOM well] ‘He
did not like that very much’ (p. 149, cited from Egil’s Saga). Thus, according
to Jónsson’s analysis the dative argument “honum” is the subject, and
nominative “það” the object.

In Chapter 7, Na’ama Pat-El studies ‘The diachrony of non-canonical subjects
in Northwest Semitic’ (155–180). The languages under investigation here are
Aramaic and Hebrew, two languages with relatively long written traditions and
a history of extensive contact both with unrelated languages and other members
of the Semitic family. However, since the non-canonical subject construction
in question seems to be a comparatively recent Northwest Semitic innovation
which cannot be reconstructed for Proto-Semitic, this allows Pat-El to trace
its development in the Hebrew and (to a lesser extent) Aramaic sources. The
predicates occurring in the construction are adjectives or stative verbs such
as Biblical Hebrew “ṭôb” ‘(be) good’, “ḥārā” ‘burn’, and “noāḥ” ‘(be)
comfortable’. (Curiously, all the Aramaic predicates seem to express modal
meanings, but one gets the impression that this may have to do with the
surviving texts, cf. p. 160.) The non-canonical subject argument is marked
with the morpheme “l(ǝ)-”, which has a number of different functions. Pat-El
argues that the one that gave rise to the non-canonical subject construction
was the use of “l(ǝ)-” to mark non-obligatory ‘free datives’, usually with
human referents, which were reinterpreted as subjects because of their high
prominence. She goes on to discuss the possible role of contact with European
languages in the maintenance and spread of the construction in Modern Hebrew.

Chapter 8, by Serena Danesi and Jóhanna Barðdal, is entitled ‘Case marking of
predicative possession in Vedic: The genitive, the dative, the locative’
(181–212). This contribution looks at the various ways of expressing
possession in one of the oldest surviving texts in an Indo-European language,
the Rigveda (2nd millennium BC). Vedic Sanskrit does not have a ‘have’ verb,
but instead expresses possession by means of a copular construction with the
possessed in the nominative case and the possessor in the genitive, dative, or
locative (thus, e.g., ‘He has supremacy’ is “asmai kṣatráṃ” [3SG.DAT
supremacy.NOM], p. 194). Danesi & Barðdal analyse the use of these
constructions in the Rigveda and find clear functional differences between
them: Whereas the genitive construction is always used with an identifiable
possessed argument (e.g. a speech act participant or a definite noun phrase
like ‘the kingdom’ or ‘the guardians of the universe’), the dative
construction is typically used when the possessed argument is unidentifiable
and non-specific (e.g. ‘treasures’, ‘greatness’, ‘any desirable good’). The
locative construction is marginal and restricted to cases where the possessor
can be conceptualised as the metaphorical location of the possessed argument.
On the basis of these findings the authors argue against claims in the
literature that the dative construction was the original predicate possession
construction in Indo-European and instead classify it as an experiencer
construction which had ‘abstract possession’ as one of its (secondary) uses.

In Chapter 9, Tonya Kim Dewey and Stephen Mark Carey diagnose a case of
‘Accusative sickness? A brief epidemic in the history of German’ (213–237).
The topic here is a small group of predicates in Old and Middle High German
where the experiencer argument appears to have changed from dative to
accusative. This is unexpected in light of the findings reported by Dunn et
al. (2017), who observe a strong cross-Germanic tendency for the change
accusative > dative (‘dative sickness’) in non-canonical subjects, not dative
> accusative. Dewey & Carey give an overview of the predicates found with
unexpected accusative subjects in the Old and Middle High German sources,
which include verbs meaning ‘think’, ‘fear’, ‘be well’, and ‘feel nausea’.
They argue that the variation between accusative and dative subjects was not
random but meaningful, the accusative being preferred for subjects with a
higher degree of affectedness. Some of the predicates investigated survive
with the dative subject construction in Modern German, the verb “dünken”
‘think’ also with the accusative. The modern variation in this verb, however,
is dialectal or idiolectal, and hence has nothing to do with the affectedness
of the experiencer argument.

Chapter 10, the first contribution to the ‘Afterword’, is Andrej Malchukov’s
‘Forty years in the search of a/the subject’ (241–256). Malchukov gives a
concise summary of some of the main thoughts and debates about the notion of
‘subject’ since the recognition of ergativity in the 1970s and the challenges
this phenomenon posed to traditional assumptions about subjecthood. The first
section deals with the various approaches to the identification or definition
of subject properties, such as Keenan’s (1976) ‘multifactorial’ approach,
Croft’s (2001) ‘Subject Construction Hierarchy’, and Lazard’s (1998)
distinction between ‘predication subjects’ and ‘reference subjects’. The
second section concerns the diachronic development of non-canonical subjects,
in particular how ‘transimpersonal’ constructions where the object is more
prominent than the subject may come to be reanalysed as intransitives with a
non-canonical subject (e.g. a Ket example meaning ‘I spin around’, lit. ‘it
spins me’). Finally, the third section provides some thoughts on the notion of
‘oblique’ subject, including Barðdal & Eythórsson’s suggestions in Chapter 11,
and notes a few remaining challenges for typologists in dealing with these
issues.

In Chapter 11, the last paper, Jóhanna Barðdal and Thórhallur Eythórsson ask
the question ‘What is a subject’ and critically examine ‘The nature and
validity of subject tests’ (257–273). In particular, Barðdal & Eythórsson
question approaches which define subjecthood with reference to (one or more)
behavioural tests and a number of more recent works which take ‘subject’ to be
a gradient notion. On the one hand, there is no general agreement about which
behavioural tests are necessary and sufficient, making comparison of different
analyses difficult and potentially involving ‘methodological opportunism’
(Croft 2001) where the criteria used are those confirming the linguist’s a
priori assumptions. On the other hand, gradient notions are criticised for not
tackling the issue properly but instead setting up poorly defined intermediate
categories (‘semi-subject’, ‘pseudo-subject’, etc.). The authors’ own proposal
is to define ‘subject’ as “the first argument of the argument structure”
(263), which in turn depends on the semantics of the predicate. They
illustrate this by examining the behaviour of non-nominative subjects in
Icelandic and German, such as Icelandic “mér varð óglatt” and German “mir
wurde übel” [me.DAT became queasy] ‘I felt queasy’. On the basis of two
subject tests, involving conjunction reduction and control, it has been argued
that the dative argument is a subject in Icelandic but not in German. Under
Barðdal & Eythórsson’s definition, both are clearly subjects. Their different
syntactic properties may be used to describe the differences between Icelandic
and German subjects, but should not be used to define these categories.

EVALUATION

This is a fine volume with many excellent and well-written contributions. The
authors’ arguments and the evidence from the various languages are presented
in a clear and readable manner, making it easy to follow the discussion even
if the language(s) under investigation is/are unfamiliar to the reader; cf.,
for instance, the exemplary chapter on Tsezic affective constructions by
Comrie, Forker & Khalilova (Chapter 3), or Danesi & Barðdal’s illuminating
study of possessive constructions in Vedic Sanskrit (Chapter 8). The volume as
a whole is coherent insofar as all contributions concern some aspect of
non-canonical subjects, but it would have been interesting to have some of the
authors respond directly to each other’s contributions, in particular those
with different positions on the notion of (non-canonical) subject. (This
could, for instance, be done in the form of short reply articles, as in Penke
& Rosenbach 2007.) Having a head-on discussion about the necessity of the
subject category might also have made the more theoretical contributions by
Farrell & Willgohs and Van Valin (Chapters 4 and 5) fit somewhat less
uncomfortably in the volume, which is otherwise devoted primarily to empirical
investigations. These two papers, which both argue against the value of the
subject notion from a RRG perspective, depart from the received opinion and
that of most of the other contributions, where ‘subject’ is clearly considered
a necessary (or at least very useful) cross-linguistic concept; see e.g. the
contributions by Jónsson (Chapter 6), Pat-El (Chapter 7), or Comrie, Forker &
Khalilova. Most explicit is Barðdal & Eythórsson’s position in their
concluding discussion; they argue that a cross-linguistically applicable
subject notion “can be defined in terms of argument structure and the relation
between the arguments” and “prompts us to delve deeper and search for
explanations for why a specific category of subjects shows deviant behavior”
(268). I find their arguments quite convincing, but it would have been
interesting to hear what Farrell & Willgohs or Van Valin would make of them.

I am rather less convinced by Dewey & Carey’s account of accusative subjects
in earlier High German. While their hypothesis that the observed
accusative/dative variation was principled is reasonable enough, I do not
think their paper presents sufficient evidence for it. For two of the six Old
High German predicates discussed (“angusten” ‘fear’ and “(un)uuillon” ‘feel
nausea’), a semantic and pragmatic comparison proves impossible because the
examples of accusative subjects are not attested in running text. For two
other verbs (“girinnen” ‘lack’ and “gilimphan” ‘be obliged’), they give only a
single example of an accusative subject, and the one cited for “gilimphan” is
from a notoriously slavish translation, the Old High German Tatian. In the
relevant clause the Vulgate has an accusative argument as well (“oportebat
autem eum transire”; John 4:4), but the authors no not at all discuss the
possibility of Latin influence on the Old High German text. In general, the
sources do not appear to have been scrutinised very carefully. The two
examples of “angusten” are said to have no source in Hennig (1957), but one
merely has to look up the verb in the standard dictionary (AWb, s.v. angusten,
sense 3) to find the reference. Incidentally, the two examples, one with a
dative and one with an accusative subject, are from different glosses to the
same passage (1 Kings [1 Samuel] 28; Steinmeyer & Sievers 1879: 405),
suggesting that the variation may not be as principled as the authors would
have. In any event, what is clear from their contribution is that there are
still many unanswered questions about non-canonical subjects in the early
Germanic languages.

The linguistic examples are generally transparent and easy to follow
throughout the volume, but it would have been convenient for the reader if the
contributors either adhered to the same glossing principles, or at least put
their abbreviations in the same place so that one does not have to search for
them; some authors give the glossing abbreviations in an appendix, some in a
footnote, others in the acknowledgements or not at all. Even more inconvenient
is when contributors state that they follow the Leipzig Glossing Rules, but
then do not adhere to them. In Pat-El’s chapter I noted about ten
abbreviations which are not in the Leipzig conventions. To be sure, one can
usually guess what the abbreviations are supposed to mean (e.g. ‘DO’ and
‘ACT’), but the point of the conventions, after all, is to make guessing
unnecessary. And not all of the glosses are self-explanatory. I was not able
to decipher ‘CP’ and ‘DUBJ’, and it required some deliberation to decide
whether ‘PART’ meant ‘partitive’ or ‘particle’. ‘P’ means ‘patient’ in the
Leipzig rules but is evidently used for ‘plural’ here, and the distinction
between - for affixes and = for clitics is not observed.

Otherwise the editing is well done and there are relatively few typos in the
volume, but a number of inconsistencies may prove distracting to the reader.
In the introduction, the reference to canonical subjects in Tsezic should be
to ‘ergative’, not ‘nominative’, subjects (p. 10). A rather confusing passage
in the chapter on Vedic Sanskrit mentions a feature [±specific] which is not
found in the table referred to (Table 2, p. 189); the passage appears to have
been copied from the description of Table 3, which does include the feature
[±specific], but one only realises this on p. 196. There are also a few
repetitive passages in the literature review on Icelandic and German on pp.
261–262. And, lastly, to give credit where it is due: The editors of the
volume are of course Barðdal, Pat-El & Carey – not Barðdal & Eythórsson, as
suggested on p. 251.

REFERENCES

AWb = Althochdeutsches Wörterbuch. 1952–. Published by the Sächsische Akademie
der Wissenschaften zu Leipzig. http://awb.saw-leipzig.de/ (1 Jun 2019).

Croft, William. 2001. Radical Construction Grammar: Syntactic theory in
typological perspective. Oxford: Oxford University Press.

Dunn, Michael, Tonya Kim Dewey, Carlee Arnett, Thórhallur Eythórsson & Jóhanna
Barðdal. 2017. Dative sickness: A phylogenetic analysis of argument structure
evolution in Germanic. Language 93(1). e1–e22.
https://doi.org/10.18148/hs/2017.v0i0.14.

Hennig, Joachim Dieter. 1957. Studien zum Subjekt impersonal gebrauchter
Verben im Althochdeutschen und Altniederdeutschen unter Berücksichtigung
gotischer und altwestnordischer Zeugnisse. Göttingen: Georg-August-Universität
Göttingen dissertation.

Keenan, Edward L. 1976. Towards a universal definition of subject. In Charles
N. Li (ed.), Subject and topic, 303–333. New York: Academic Press.

Lazard, Gilbert. 1998. Actancy (Empirical Approaches to Language Typology 19).
Berlin: De Gruyter Mouton. https://doi.org/10.1515/9783110808100.

Penke, Martina & Anette Rosenbach (eds.). 2007. What counts as evidence in
linguistics: The case of innateness (Benjamins Current Topics 7). Amsterdam:
John Benjamins. https://doi.org/10.1075/bct.7.

Steinmeyer, Elias & Eduard Sievers. 1879. Die althochdeutschen Glossen. Erster
Band: Glossen zu biblischen Schriften. Berlin: Weidmann.
http://digital.ub.uni-duesseldorf.de/content/titleinfo/289355 (1 Jun 2019).

Thráinsson, Höskuldur. 2007. The syntax of Icelandic (Cambridge Syntax
Guides). Cambridge: Cambridge University Press.
https://doi.org/10.1017/CBO9780511619441.


ABOUT THE REVIEWER

Sune Gregersen is a PhD student at the Amsterdam Center for Language and
Communication (University of Amsterdam), working on modality in the older
Germanic languages. His other interests include the early history of
linguistic description, tense–mood–aspect across languages, and the diachrony
of (non-canonical) argument marking.





------------------------------------------------------------------------------

***************************    LINGUIST List Support    ***************************
 The 2019 Fund Drive is under way! Please visit https://funddrive.linguistlist.org
  to find out how to donate and check how your university, country or discipline
     ranks in the fund drive challenges. Or go directly to the donation site:
               https://iufoundation.fundly.com/the-linguist-list-2019

                        Let's make this a short fund drive!
                Please feel free to share the link to our campaign:
                    https://funddrive.linguistlist.org/donate/
 


----------------------------------------------------------
LINGUIST List: Vol-30-3270	
----------------------------------------------------------