ArticlePDF Available

The perfective form in Arusa - A cognitive-grammaticalization model

Authors:

Abstract and Figures

The present paper studies the semantics of the so-called perfective (PFV) form in Arusa (Maasai), using the model of the dynamic (one- and two-dimensional) semantic maps. The analysis demonstrates that PFV is a broad, semi-advanced resultative-path gram. It spans large sections of the two sub-paths of the resultative path: the anterior path (present perfect, perfective and non-perfective past, as well pluperfect and future perfect) and the simultaneous path (present stative). However, the PFV form is incompatible with the input sense of the resultative path (a resultative proper present) and the most advanced stages of the two sub-paths (non-stative present and progressive past). If the information related to prototypicality is included, the map adopts the shape of a wave with the prototypicality peaks located in the area of a perfective past and, to a lesser degree, a present perfect. The senses of a non-perfective past and a stative present are less prototypical. Other senses (pluperfect, future perfect and counterfactual irrealis) are non-prototypical, contributing minimally to the gram’s semantics.
Content may be subject to copyright.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
69
THE PERFECTIVE FORM IN ARUSA
– A COGNITIVE-GRAMMATICALIZATION MODEL
Alexander ANDRASON
Department of African Languages, University of Stellenbosch
Matieland, 7602, South Africa
andrason@sun.ac.za
Michael KARANI
Department of Foreign Languages and Linguistics, University of Dar es Salam
Department of African Languages, University of Stellenbosch
Dar es Salam, Tanzania
karanim@udsm.ac.tz
The present paper studies the semantics of the so-called perfective (PFV) form in Arusa (Maasai),
using the model of the dynamic (one- and two-dimensional) semantic maps. The analysis
demonstrates that PFV is a broad, semi-advanced resultative-path gram. It spans large sections of
the two sub-paths of the resultative path: the anterior path (present perfect, perfective and non-
perfective past, as well pluperfect and future perfect) and the simultaneous path (present stative).
However, the PFV form is incompatible with the input sense of the resultative path (a resultative
proper present) and the most advanced stages of the two sub-paths (non-stative present and
progressive past). If the information related to prototypicality is included, the map adopts the
shape of a wave with the prototypicality peaks located in the area of a perfective past and, to a
lesser degree, a present perfect. The senses of a non-perfective past and a stative present are less
prototypical. Other senses (pluperfect, future perfect and counterfactual irrealis) are non-
prototypical, contributing minimally to the gram’s semantics.
Keywords: Maasai, TAM verbal semantics, maps, cognitive linguistics, grammaticalization
1. Introduction
Arusa is an under-researched eastern Maasai language of the Nilotic family,
spoken in Tanzania in the area near the Mount Kilimanjaro. The present paper
deals with one aspect of this language, namely the semantics of the “Perfective”
Asian and African Studies, Volume 26, Number 1, 2017
70
gram(mmatical construction), henceforth, referred to as PFV.1 In our analysis,
we use the model of a two-dimensional dynamic map (the so-called wave
model), which has been developed at the interfaces of cognitive linguistics and
grammaticalization theory (for detail, see section 2. 2).2
In order to achieve this goal, the paper will be organized as follows: To begin
with, in section 2, we will present the background of our study and familiarize
the reader with the framework underlying this research. Subsequently, in
section 3, we will introduce empirical evidence. We will illustrate the various
senses exhibited by the PFV form, discussing the context of their use and
relationships to other meanings and forms found in the Arusa language. In
section 4, this evidence will be analyzed within the adopted theoretical
framework. First, the meaning of the PFV gram will be explained as a wave.
Second, this definition will be contrasted with (a few) available diachronic
and/or comparative facts. Lastly, in section 5, main conclusions will be drawn
and lines of future research proposed.
2. Preliminaries
2.1. PFV – Its Formal Characteristics and Scholarly Tradition
The PFV form is marked in a variety of manners in Arusa. Two principal types
of marking depend on the class of verb. In Class I, PFV is typically encoded by
two morphemes, namely the affix -tV- and a vowel suffix, e. g. a-ti-sir-a
empalai ‘I wrote the letter’ or a-ta-bol-o ‘I opened’.3 In most cases, the vowel
of the affix -tV- is analogous to the vowel of the stem. However, exceptions
1 The use of this label partially frees us from equalling the form in question with any
meaning. That is, even though the label PFV may be seen to be an abbreviation of
‘perfective’ (thus referring to an aspect), it should be taken as a mere symbol with no
intrinsic semantic content.
2 ANDRASON, A. A Complex System of Complex Predicates: Tense, Taxis, Aspect and
Mood in Basse Mandinka from a Grammaticalization and Cognitive Perspective;
ANDRASON, A. From Vectors to Waves and Streams: An Alternative Approach to
Semantic Maps. In Stellenbosch Papers in Linguistics, 2016, Vol. 45, pp. 129.
3 The morpheme -tV- has a wider application in Arusa, being also found in the
Subjunctive, Imperative and Infinitive, see LEVERGOOD, B. J. Topics in Arusa
Phonology and Morphology, p. 35. For a similar situation in other Maasai varieties
consult: TUCKER, A. N., MPAAYEI, J. T. O. A Grammar of Maasai with Vocabulary;
KONIG, C. Aspekt im Maa; KOOPMAN, H. On the Homophony of Past Tense and
Imperatives in Kisongo Maasai. In UCLA Working Papers in Linguistics, 2001, Vol. 6,
pp. 113.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
71
occur, as demonstrated by the verb rem ‘poke, stub’, which exhibits the
element -ta-, átaremo ‘I poked’. The suffix can be either one of the three
vowels -a, -o or -e (as in e-irrag-a ‘he slept’, e-ipot-o ‘he called him’ and e-
irug-e ‘he bent’) or a diphthong (-yie or yio; for instance e-ita-yio ‘he
removed’ and e-iŋura-yie ‘he looks at him’). The distribution of the vocalic
variants often depends on the quality of the radical vowel. The suffix -a is
found after the radical vowels a or u, the suffix -o appears after o; and the suffix
-e is used after i.4 However, a considerable number of counter-examples to
these tendencies are also found. The suffixes -yie and -yio are employed if the
root ends in a long vowel or a diphthong, e. g. -aa and -ai. Class II is only
marked by suffixes. These suffixes can, again, take form of a vowel (-a, -o or -
e) or diphthongs (-yie or -yio). Their distribution is identical to that found in
Class I. Additionally, in the Middle and Reflexive constructions, the regular
marker of PFV is the suffix -e which can complement the morpheme -tV-
(Class I) and/or replace the suffixes -a, -o, -yie or -yio, for instance, e-ta-bol-e
‘it opened’, e-iŋol-e ‘it mixed’, and a-isuj-e ‘I washed.5 There is also a group of
irregular verbs that use suppletive forms to mark the PFV form: a-lo ‘I go’ – a-
ʃomo ‘I went’ or a-ɲa ‘I eat’ – ataama ‘I ate’.6
The semantics of the PFV form in Arusa has not been researched in depth.
The gram has been briefly described in two studies dedicated to the Arusa
language, being defined as a past tense 7 or, more recently, as a perfective
aspect.8 In another Tanzanian variety of Maasai, Kisongo, the equivalent form
has been viewed as a past tense of a participial origin.9 The understanding of
PFV in Kenyan (more standard) Maasai and its dialects is similar. Tucker and
Mpaayei argue that PFV is a past tense that corresponds to two verbal
4 Compare with TUCKER, A. N., MPAAYEI, J. T. O. A Grammar of Maasai with
Vocabulary, p. 56.
5 Compare PAYNE, D., OLSEN, D. Maa (Maasai) Nominalization: Animacy,
Agentivity and Instrument. In MATONDO, M. et al. (ed.). Selected Proceedings of the
38th Annual Conference on African Linguistics. Somerville: Cascadilla Proceedings
Project, 2009, p. 155.
6 KARANI, M. The Syntactic Categories and Argument Structure in the Verbal
Complex. For a detailed description of formal properties of PFV, in other Maasai
varieties consult TUCKER, A. N., MPAAYEI, J. T. O. A Grammar of Maasai with
Vocabulary; KONIG, C. Aspekt im Maa; KOOPMAN, H. On the Homophony of Past
Tense and Imperatives in Kisongo Maasai. In UCLA Working Papers in Linguistics,
2001, Vol. 6, pp. 113.
7 LEVERGOOD, B. J. Topics in Arusa Phonology and Morphology.
8 KARANI, M. The Arusa Verb System.
9 KOOPMAN, H. On the Homophony of Past Tense and Imperatives in Kisongo
Maasai. In UCLA Working Papers in Linguistics, 2001, Vol. 6, pp. 113.
Asian and African Studies, Volume 26, Number 1, 2017
72
constructions in English: the Past and the Perfect.10 In a study of the Camus
dialect of Kenya, Konig defines PFV as a perfective aspect and the Maasai
language in general as aspect-prominent.11 Konig devotes much attention to the
interaction between the PFV morpheme(s) and verbal Aktionsart, and identifies
the following correlations: in PFV, action verbs express a completed action;
telic verbs indicate the final or terminative stage of the action; and inchoative-
stative verbs acquire an inchoative sense. In contrast, “totally stative” verbs are
incompatible with the PFV morpheme. The aspectual view proposed the Camus
dialect has posteriorly been extrapolated to other, typically Kenyan, Maasai
varieties12the Maasai being viewed as an aspectual language.13
2.2. Framework
In the present study, in order to analyze and explain the meaning of the PFV
form in Arusa, we will use the model of a dynamic semantic map.14
10 TUCKER, A. N., MPAAYEI, J. T. O. A Grammar of Maasai with Vocabulary, p. 61.
This position may already be found in HOLLIS who identifies PFV with the English
Perfect and Past, see HOLLIS, A. C. The Masai. Their Language and Folklore, p. 58.
11 KONIG, C. Aspekt im Maa.
12 See PAYNE, D., HAMAYA, M., JACOBS, P. Active, Passive, and Inverse in Maasai.
In GIVÓN, T. (ed.). Voice and Inversion, 283315; HAMAYA, M. Vowel Harmony in
Maasai. In Linguistics, 1997, Vol. 607, pp. 130; PAYNE, D., OLE-KOTIKASH, L.,
OLE-LEKUTIT, K. M. A Frame Semantics Approach to Lexemic Structure:
Uncovering the Truth about Maa a-síp. In Journal of African Languages and Linguistics,
2001, Vol. 22, pp. 123; PAYNE, D., OLSEN, D. Maa (Maasai) Nominalization:
Animacy, Agentivity and Instrument. In MATONDO, M. et al. (ed.). Selected
Proceedings of the 38th Annual Conference on African Linguistics. Somerville:
Cascadilla Proceedings Project, 2009.
13 In a narrative discourse, the so-called N-tense is used in the past when describing
events in a sequence, see TUCKER, A. N., MPAAYEI, J. T. O. A Grammar of Maasai
with Vocabulary, p. 61. This form consists of an Imperfective (IPFV) headed by the
relativizer n-. Such sequential events may be perfective or perfectal, e.g. Etaara ɔlarro,
neyieŋ, neya aaŋ ‘He killed a buffalo, slaughtered it, and took home’. However, N-
tense may apply to all time frames, including present and future. The temporal
interpretation of the N-tense usually depends on the reference time of the first verb in
the sentence.
14 HASPELMATH, M. The Geometry of Grammatical Meaning Semantic Maps and
Cross-linguistic Comparison. In THOMASELLO, M. (ed.). The New Psychology of
Language, pp. 211242; HASPELMATH, M. Coordinating Constructions: An
Overview. In HASPELMATH, M. (ed.). Coordinating Constructions, pp. 140;
JANDA, L. Cognitive Linguistics in the Year 2015. In Cognitive Semantics, 2015, Vol.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
73
Semantic maps build on the fact that grams are synchronically polysemous
and that this synchronic polysemy is internally coherent, all the senses being
connected historically and conceptually. Historically, each sense emerges
gradually from another sense and gives rise to yet another sense. Conceptually,
the above-mentioned emergence of a sense from another sense is always
motivated by human cognitive mechanisms, such as metonymy, metaphor,
context induced reinterpretation, and analogy.15 The graphic portrayal of such
historical and conceptual connections delivers a map. In this map, each node
represents a sense and each linking line symbolizes its historical and conceptual
foundation.16
It is evident that this type of map has an inherent dynamic dimension the
organization of the map reproduces the grammaticalization path that underlies
the evolution of the form. The map defines the meaning of a form as a portion
of a grammaticalization cline or a cluster of such paths. Depending on the level
of granularity, it can appear as a cline or as a network. In this map, each
component (a sense available synchronically) is correlated with a stage on the
grammaticalization path along which the form has been travelling.17
Frequently, due to the pervasiveness of the certain grammaticalization paths,
the range of senses exhibited synchronically by a gram is sufficient to postulate
a map. That is, the map is posited because the senses offered by a form match
one of the (nearly universal) grammaticalization templates.18 This property is
1, pp. 131154; ANDRASON, A. A Complex System of Complex Predicates: Tense,
Taxis, Aspect and Mood in Basse Mandinka from a Grammaticalization and Cognitive
Perspective; ANDRASON, A. From Vectors to Waves and Streams: An Alternative
Approach to Semantic Maps. In Stellenbosch Papers in Linguistics, 2016, Vol. 45, pp.
1–29.
15 JANDA, L. Cognitive Linguistics in the Year 2015. In Cognitive Semantics, 2015,
Vol. 1, pp. 131154.
16 HASPELMATH, M. The Geometry of Grammatical Meaning Semantic Maps and
Cross-linguistic Comparison. In THOMASELLO, M. (ed.). The New Psychology of
Language, pp. 211242; HASPELMATH, M. Coordinating Constructions: An
Overview. In HASPELMATH, M. (ed.). Coordinating Constructions, pp. 140.
17 HASPELMATH, M. The Geometry of Grammatical Meaning Semantic Maps and
Cross-linguistic Comparison. In THOMASELLO, M. (ed.). The New Psychology of
Language, pp. 211242; HASPELMATH, M. Coordinating Constructions: An
Overview. In HASPELMATH, M. (ed.). Coordinating Constructions, pp. 140;
ANDRASON, A. A Complex System of Complex Predicates: Tense, Taxis, Aspect and
Mood in Basse Mandinka from a Grammaticalization and Cognitive Perspective;
ANDRASON, A. From Vectors to Waves and Streams: An Alternative Approach to
Semantic Maps. In Stellenbosch Papers in Linguistics, 2016, Vol. 45, pp. 129.
18 HEINE, B., CLAUDI, U., HÜNNEMEYER, F. Grammaticalization. A Conceptual
Framework; HEINE, B. Cognitive Foundations of Grammar; HASPELMATH, M. The
Asian and African Studies, Volume 26, Number 1, 2017
74
particularly useful in cases where direct diachronic evidence, which could
corroborate the grammaticalization path travelled by a gram, is scarce or
unavailable. Nevertheless, direct or indirect diachronic proofs can importantly
validate a map and/or render it more precise. This verification becomes crucial
in instances where more than one evolutionary scenario (and hence more than
one map) is possible. Overall, a map is always “gram specific”. That is, even
though it is based on typological templates, it reflects the actual
grammaticalization cline of a form. It shows how the senses of the form have
been acquired. Therefore, divergences from canonical (nearly universal) paths
are possible.19
Overall, the semantic coherence of a form does not stem from its invariant or
abstract meaning, which is subsequently modulated in specific contexts where
that form is used. Senses available synchronically do not emerge from a shared
semantic core. The semantic coherence rather reflects the diachronic reiteration
of the meaning-extension process.20 The unity of a gram lies in the path the
gram’s meaning being the map itself. In other words, a gram exhibits
synchronically a range of senses that co-exist simultaneously along the path.
Some of them were originally pragmatic meaning extensions that have
gradually been semanticized, and that have prompted further extensions. 21
Crucially, all the senses (i.e. components of the map) are connected to each
other through family resemblance. This means that even though each sense
shares some properties with its predecessor and successor, two distant senses
may not share any properties at all, or the properties they share may be trivial.22
Therefore, although the map is coherent in its totality, the actual semantic
Geometry of Grammatical Meaning Semantic Maps and Cross-linguistic Comparison.
In THOMASELLO, M. (ed.). The New Psychology of Language, pp. 211242.
19 ANDRASON, A. El sistema verbal hebreo en su contexto semítico: una visión
dinámica [The Hebrew Verbal System in its Semitic Context: A Dynamic Perspective];
ANDRASON, A. A Complex System of Complex Predicates: Tense, Taxis, Aspect and
Mood in Basse Mandinka from a Grammaticalization and Cognitive Perspective.
20 BYBEE, J. Language, Usage and Cognition.
21 ANDRASON, A., LOCATELL, C. The Perfect Wave. In Biblical and Ancient Greek
Linguistics, 2016, Vol. 5, pp. 7122.
22 COUSSÉ, E. Lexical Expansion in the HAVE and BE Perfect in Dutch A
Constructionist Prototype Account. In Diachronica, 2014, Vol. 31, pp. 159191;
COUSSÉ, E. Grammaticalization, Host-class Expansion and Category Change. In
NORDE, M., VAN GOETHEM, K., COUSSÉ, E., VANDERBAUWHEDE, G. (eds.).
Category Change from a Construction Grammar Perspective; JANDA, L. Cognitive
Linguistics in the Year 2015. In Cognitive Semantics, 2015, Vol. 1, pp. 131154;
ANDRASON, A., LOCATELL, C. The Perfect Wave. In Biblical and Ancient Greek
Linguistics, 2016, Vol. 5, pp. 7122.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
75
similarity only involves adjacent senses. For two distant senses, there may be no
direct conceptual relation. Their connection emerges from the reiteration of
conceptual relations of adjacent senses.23 This implies, in turn, that the increase
of distance on the path is inversely proportional to cognitive proximity (and/or
derivability) – adjacent senses can be related psychologically by speakers while
distant senses cannot.24
Since a map represents senses as consecutive stages emerging from one
another, it is directional and can be understood as a vector. This vector informs
one about the direction of diachronic meaning extensions. It tells us not only
what the form is but also by which types of dynamics it is governed: where it
came from and where it is heading to. The synchronic state ceases to be static,
but becomes dynamic. Synchrony is no longer equivalent to stasis it has
become a process.25
Most maps, reshaped as vectors, are qualitative and pay equal importance to
all the components, i.e. senses or stages. However, although a form exhibits
many senses, not all of them are equally relevant. Some are prototypical, others
are not. The concept of a vector is, therefore, sometimes expanded to
accommodate some type of quantitative information related to prototypicality.26
That is, the horizontal x axis representing the vector of meaning-extension is
accompanied by the vertical y axis that specifies the degree of prototypicality.
23 ANDRASON, A., LOCATELL, C. The Perfect Wave. In Biblical and Ancient Greek
Linguistics, 2016, Vol. 5, pp. 7122.
24 In the case of distant senses, any attempt to determine an invariant or shared meaning
is unsatisfactory. An archetypal example can be the forms ran (Past tense) and can
(Present tense of Preterite-Present verbs) in Germanic languages, which both derive
from the same morphology, still visible in Modern Icelandic. For contemporary
speakers of English there is no relation between these two senses and forms even
though they evolved from the same input. In English, the conceptual distance is so
extensive that the Preterite-Present verbs (which are morphologically Past tenses) have
been included in the paradigm of the Present tense. The invariant and/or shared meaning
of these forms (if postulated) would be trivial, see ANDRASON, A., LOCATELL, C.
The Perfect Wave. In Biblical and Ancient Greek Linguistics, 2016, Vol. 5, pp. 7122.
25 ANDRASON, A. A Complex System of Complex Predicates: Tense, Taxis, Aspect
and Mood in Basse Mandinka from a Grammaticalization and Cognitive Perspective;
ANDRASON, A. From Vectors to Waves and Streams: An Alternative Approach to
Semantic Maps. In Stellenbosch Papers in Linguistics, 2016, Vol. 45, pp. 129.
26 ANDRASON, A. A Complex System of Complex Predicates: Tense, Taxis, Aspect
and Mood in Basse Mandinka from a Grammaticalization and Cognitive Perspective;
ANDRASON, A. From Vectors to Waves and Streams: An Alternative Approach to
Semantic Maps. In Stellenbosch Papers in Linguistics, 2016, Vol. 45, pp. 1–29;
ANDRASON, A., LOCATELL, C. The Perfect Wave. In Biblical and Ancient Greek
Linguistics, 2016, Vol. 5, pp. 7122.
Asian and African Studies, Volume 26, Number 1, 2017
76
When the arguments of the former are correlated with the values of the latter,
the model of meaning receives a two-dimensional form of a wave. The wave
specifies not only the range of semantic potential and its coherence (as the
vector does) but also the prototypical zone visualized as the wave peak. This
peak indicates the most semanticized and cognitively active part of the semantic
potential. The senses located outside the wave peak are non-prototypical.
Usually, they are (more) pragmatically driven or archaic.27
Certainly, prototypicality constitutes a difficult concept and depends on
various factors: frequency, saliency, productivity, constraints, relations to other
forms and their meanings, etc. Frequency itself is complex, as it usually gives
different results in different types of text, genres or contexts. It can also be
driven semasiologically (what sense of a form is the most frequent?) or
onomasiologically (what form is the most common expression of a sense?). In
this study, the degree of prototypicality will be deduced from native speakers’
perception (which sense is first to come to mind?), the sense’s productivity
(compatibility with roots) and the absence of constraints (acceptability in all
typologically possible contexts in which a given sense may be activated).28
Overall, the procedure adopted in this paper will consist of the following: 1)
the identification of all possible senses the PFV form can express and the
determination of their prototypicality; 2) the organization of these senses into a
map that is compatible with a common grammaticalization path (or a set of
related paths) and, by introducing the information related to prototypicality, the
presentation of this map as a wave; and lastly 3) corroboration of the bi-
dimensional map by certain diachronic and comparative fact.
27 The idea of a wave makes reference to the dynamics underlying the form as it travels
along the grammaticalization path. The form advances not only with respect to the
scope of its semantic potential (it becomes compatible with new senses), but also with
respect to its prototypicality (further senses of the path become prototypical).
ANDRASON, A. A Complex System of Complex Predicates: Tense, Taxis, Aspect and
Mood in Basse Mandinka from a Grammaticalization and Cognitive Perspective;
ANDRASON, A. From Vectors to Waves and Streams: An Alternative Approach to
Semantic Maps. In Stellenbosch Papers in Linguistics, 2016, Vol. 45, pp. 129;
ANDRASON, A., LOCATELL, C. The Perfect Wave. In Biblical and Ancient Greek
Linguistics, 2016, Vol. 5, pp. 7122.
28 Compare ANDRASON, A. A Complex System of Complex Predicates: Tense, Taxis,
Aspect and Mood in Basse Mandinka from a Grammaticalization and Cognitive
Perspective. This approach mainly stems from the absence of written texts and larger
corpora in the Arusa language. This methodological limitation may soon be overcome
as the authors of this paper currently develop a research project specifically dedicated to
the documentation of oral literature in Arusa.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
77
3. Evidence
The PFV form in Arusa is compatible with a number of perfectal (or taxis)
senses, i.e. senses that are cross-linguistically associated with the category of a
perfect. The most important and prototypical class of such values includes
senses that belong to the semantic domain of a present perfect.
The PFV gram commonly expresses the value of a present perfect resultative.
In this usage, it indicates that a dynamic event has already occurred and its
effects are highly relevant for the current state of affairs. 29 This can be
illustrated by (1. a) and (1. b) where the results of the two anterior actions (i.e.
losing the keys and breaking the arm) persist unchanged – the keys are still lost
and the arm is still unhealed.
(1) a. ma-idim aatijiŋa aatua engaji ai, a-imin-ie
ifungooni
NEG.1SG-able INF.enter inside house my 1SG-loose-PFV
PL.key
I cannot enter my house. I have lost the keys.
b. mi-kii-ŋgum. A-ti-gil-a enkaina ai tenakata
NEG-2>1-touch. 1SG-PFV-break-PFV arm my now
Do not touch me. I have just broken my arm.
Another perfectal sense compatible with PFV is the value of a present
perfect experiential (see examples 2. a-b below). This sense portrays an action
as an experience that has occurred at least once in the span of time ranging from
the past till the present. The action itself occurred before the speech time,
possibly long time ago. The exact time of its occurrence cannot, however, be
specified overtly. The nuance of current relevance persists, although it relates to
the general experience of the agent, rather than to the action itself. The
resultative inferences of that anterior event are absent – the reality might have
changed since the moment the event occurred.30
29 MCCAWLEY, J. Tense and Time Reference in English. In FILLMORE, Ch. J.,
LANGENDOEN, D. T. (eds.). Studies in Linguistics and Semantics, pp. 96113;
BYBEE, J., PERKINS, R., PAGLIUCA, W. The Evolution of Grammar, pp. 6162.
30 COMRIE, B. Aspect: An Introduction to the Study of Verbal Aspect and Related
Problems; DAHL, Ö. Tense and Aspect Systems, pp. 140–142; BYBEE, J., PERKINS,
R., PAGLIUCA, W. The Evolution of Grammar, p. 62; DE HAAN, F. Typology of
Tense, Aspect, and Modality Systems. In SONG, J. J. (ed.). The Oxford Handbook of
Linguistic Typology, p. 457.
Asian and African Studies, Volume 26, Number 1, 2017
78
(2) a. iʃomo aikata Nairobi?
2-go.PFV ever Nairobi
Have you ever been to Nairobi?
b. a-ta-dua ina sinema. A-iŋura-yie te enda
jumaa naima
1SG-PFV-see that movie 1SG-watch-PFV in that week
past
I have seen that movie. I watched it last week.
Scholars distinguish a further subtype of present perfect, i.e. the present
perfect inclusive, also referred to as continuing. This sense communicates that
an action began in the past and continues into present time. That is, an activity
or situation holds continuously from a moment in the past to the present.31 The
present perfect inclusive is expressed in Arusa in two manners depending on
how the continuity of an action is construed. If the expression of for how long
the action has been occurring is overt, a bi-clausal construction is used. The first
clause contains the verb in PFV and the expression of duration. This clause,
however, does not necessarily imply that the action has been continuing into
present time. As will be evident from the subsequent discussion, it could also
refer to a past time frame, denoting duration in the past. Therefore, an additional
clause is needed in which the action is depicted as ongoing presently. This
current ongoing-ness is usually communicated by means of an Imperfective
(IPFV) or a Progressive (PROG) form (3. a). Additionally, the inclusive perfect
may specify the starting point of the action, i.e. where it began to occur in the
past (since when). In such instances, the IPFV or PROG forms are regularly
used (3. b-c). If PFV was employed, it would imply that the action ceased and
does not hold anymore.
(3) a. a-iteŋenw-e enkutuk oo lMaasai too larin tomon.
1SG-learn-PFV language of Maasai for years ten.
I was learning the Maasai language for ten years.
Ne-ŋorr ake a-iteŋenw-o
REL-not.yet still 1SG-learn-IPFV
And I am still learning it (i.e. I have been learning Maasai for 10 years).
31 COMRIE, B. Aspect: An Introduction to the Study of Verbal Aspect and Related
Problems; BYBEE, J., PERKINS, R., PAGLIUCA, W. The Evolution of Grammar, p.
62; NURSE, D. Tense and Aspect in Bantu, p. 154; DE HAAN, F. Typology of Tense,
Aspect, and Modality Systems. In SONG, J. J. (ed.). The Oxford Handbook of
Linguistic Typology, p. 456.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
79
b. a-isom-iʃo aiter ŋole
1SG-read-PROG from yesterday
I have been reading since yesterday.
c. a-maɲ-a ene aiteru 1990
1SG-live-IPFV here from 1990
I have been living here since 1990.
To sum up, the use of PFV in the function of a present perfect inclusive is
highly limited. The PFV gram can only appear in one of the two senses
associated with that category. Furthermore, even in the subtype in which it can
be used, PFV must obligatorily be accompanied by IPFV or PROG.
PFV can also be employed with an explicit past reference, thus functioning
as a definite past. This past reference may be of any distance from the speaker’s
now. There is virtually no limit for the range of the remoteness. That is, PFV
tolerates both the most recent and the most distant points in time. To begin with,
the past event can be temporarily recent or near, for instance hodiernal (today’s
past; 4. a) and hesternal (yesterday’s past; 4. b).32
(4) a. a-iɲaŋu-a engarim tena nasirie. E-itaʃ-o te ende
1SG-buy-PFV car this morning. 3-stand-IPFV over there
I bought a car this morning. It parks over there.
b. a-ta-leen-o elatia ai ŋole
1SG-PFV-visit-PFV neighbor my yesterday
I visited my neighbor yesterday.
However, the action expressed by PFV may also be located in a more distant
or even remote past. For instance, in example (5. a), the event occurred a year
ago, while in (5. b), it took place many hundreds of years ago (5. b). Overall,
the definite past uses of PFV are highly common, constituting one of the most
prototypical values of this form.
(5) a. a-idip-a ʃule tele ari oima
1SG-finish-PFV school this year past
I finished the school last year.
b. e-etuo ilmaasai Tanzania ilarin iip ooima
3-come.PFV Maasai Tanzania years one.hundred past
The Maasai came to Tanzania a hundred years ago.
32 BYBEE, J., PERKINS, R., PAGLIUCA, W. The Evolution of Grammar, p. 98.
Asian and African Studies, Volume 26, Number 1, 2017
80
As is already evident from the examples introduced previously, PFV
frequently exhibits a perfective nuance when it is employed in a past time frame
(see 4. a-b and 5. a-b). That is, PFV is compatible with past events depicted as
punctiliar, bounded and complete. Additionally, PFV may encode actions that
lack an internal event-structure, which is as another cross-linguistic
characteristic of a perfective domain.33
Sentences in (6. a) and (6. b) are further examples of that perfective past
sense. The past events of going to Dar es Salam and getting married are
presented as punctiliar and complete, having occurred once in a precise and
discrete moment on the time line:
(6) a. a-ʃomo Darsalam tele áβa oima
1SG-go.PFV Dar es Salaam this month past
Last month, I went to Dar es Salaam.
b. a-yam-iʃe ilarin tomon ooima
1SG-marry-PFV year ten past
I got married 10 years ago.
However, PFV may also express a range of nuances that are not punctiliar,
unique, discrete and/or bounded. In all such cases, the gram fails to behave as a
canonical perfective past. Two types of this non-perfective domain, compatible
with PFV, can be distinguished: durative and iterative-habitual.
PFV can indicate definite past activities whose duration is profiled. Even
though such actions are still bounded within a determined frame of time, they
span larger periods of time and correspond to imperfective past grams in other
languages (for instance in the Romance or Slavonic families). The duration may
itself be bounded in two manners. First, the range of time within which the
activity happened can be specified by stating its starting and ending points (i.e.
from when to when; 7. a). Second, the duration may also be specified by
expressing the temporal extent of the occurrence of the action (i.e. for how long;
7. b). It should be noted that in all such bounded durative examples, PFV does
not necessarily imply that the action was completed or terminated (e.g. that a
person actually learned Swahili completely; see examples 7. a-b), which
approximates it to the category of an imperfective past (compare a similar
behavior of the imperfective past in Romance and Slavonic).
33 This distinguishes them from perfects, which are characterized by a bi-dimensional
event-structure, i.e. prior cause (event) and posterior result (state or relevance). DAHL,
Ö. Tense and Aspect Systems; BYBEE, J., PERKINS, R., PAGLIUCA, W. The
Evolution of Grammar, pp. 54, 287; NURSE, D. Tense and Aspect in Bantu, pp. 134
135; DE HAAN, F. Typology of Tense, Aspect, and Modality Systems. In SONG, J. J.
(ed.). The Oxford Handbook of Linguistic Typology, pp. 450451.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
81
(7) a. a-iteŋen-we Kiswahili aiter 2000 me-tabaiki 2005
1SG-learn-PFV Kiswahili from 2000 SUB-till 2005
I studied Swahili from 2000 till 2005.
b. a-iteŋen-we Kiswahili tiatua ilarin tomon, amaɲa
Darsalam
1SG-learn-PFV Kiswahili in years ten, 1SG.live.IPFV
Dar es Salaam
I studied Swahili during 10 years, while I lived in Dar es Salam.
The other non-perfective class of uses of PFV concerns cases where the
event expressed by the PFV form happened repetitively and/or habitually. In
such instances, the gram conveys iterative and customary nuances. These
nuances can be made overt by employing expressions such as naleŋ ‘a lot’, kila
olapa ‘every month’ or kat kumok ‘quite often’:
(8) a. ore ara kiiɲi, a-iguran-a naleŋ/kat kumok
When 1SG.be.IPFV little, 1SG-play-PFV a lot/time many
When I was little, I played football a lot / quite often.
a. a-ta-la-a kila olapa
1SG-PFV-pay-PFV every month
I paid every month.
However, the use of PFV is ungrammatical with the word kutwa ‘always;
every day’ which can be regarded as the most prototypical exponent of
habituality. In such a context, it is the IPFV form that must be employed:
(9) a. ore ara kiiɲi, a-iguran empira kutwa
when 1SG.be.IPFV little 1SG-play.IPFV ball every day
When I was little, I played football every day.
b. ore ara kiti, áa-joki yeyio
kutwa
when 1SG.be.IPFV little 3>1-tell.DAT.IPFV my.mother
every day
When I was a kid, my mother always told me
páaku enkerai sidai
SUB.1SG.be child good
to be a good boy.
In cases where the nuance of past progressivity, or past ongoing-ness, needs
to be conveyed, once more, IPFV must be used, PFV being ungrammatical:
Asian and African Studies, Volume 26, Number 1, 2017
82
(10) a. ore ŋole too saan 5. 15,
when yesterday at hour 5. 15
Yesterday at 5. 15,
a-ilim-iʃo, ne-lotu ɔlcɔre lai
1SG-weed-PROG REL-come.IPFV friend my
while I was weeding, my friend came.
b. ore a-iŋur-ita empire te enteleviʃen,
when 1SG-watch-PROG ball on television,
While I was watching a football match,
ne-daɲa enteleviʃen
REL-break.IPFV television
the TV broke down.
The PFV form is also used as a pluperfect, which constitutes a past
equivalent of the category of a present perfect. Grosso modo, PFV is compatible
with two main subtypes of the semantic domain associated cross-linguistically
with pluperfects. First, PFV can introduce an anterior event from a past
perspective. That is, it can present a prior action from a mental space that is
already located in the past. This type is particularly common in reported speech
(11. a). Second, PFV can communicate anteriority in the past, denoting that an
event had occurred before another past action – all of this construed within a
single mental space (11. b). However, in this latter use, it is usually not the PFV
form of the main verb that is used, but rather the PFV of the verb idip ‘to finish’
followed by the Infinitive (the so-called subjunctive-like B Infinitive),34 e.g.
eidipa áatabaiki ‘he finished to visit’ in (11. c).
(11) a. e-te-jo ŋole e-tu-ŋayie
3-PFV-say yesterday 3-PFV-leave
Yesterday he said that
enkitok eɲe aiter enjumaa naima
wife his from week past
his wife had left him a week ago.
b. e-ewu-o itu a-lo
3-come-PFV NEG 1SG-go.IPFV
He had come, before I left.
c. ore eŋor e-itu e-lotu enatii ŋole,
when before 3-NEG 3-come.IPFV to.me yesterday,
Before he came to me yesterday,
34 TUCKER, A. N., MPAAYEI, J. T. O. A Grammar of Maasai with Vocabulary, pp. 66,
9899.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
83
e-idip-a áatabaiki ŋotoɲe
3-finish-PFV INF.visit his.mother
he had visited his mother.
The same analytical construction built around the verb idip ‘finish’ may be
used to convey meanings other than pluperfect discussed in the previous
paragraph. This usage corresponds to the semantic domains of a present perfect
or a recent past. Rarely (if ever), it is compatible with the degree of remoteness
that goes beyond a hesternal past. In all such uses, the construction expresses
the idea of prior completed-ness of an event its termination or total
accomplishment:
(12) a. taama endaa!
IMPR.eat food
Eat food!
a-idip-a áataama endaa!
1SG-finish-PFV INF.eat food
I have just eaten!
b. kanu i-loito aa-baiki yeyino
when 2-go.PROG INF-visit mother.your
When are you going to see your mother?
a-idip-a aatabaiki ŋole!
1SG-finish-PFV INF.visit yesterday
I visited her yesterday.
The sense of anteriority, commonly associated with the PFV gram in Arusa,
is not limited to the present and past time frames, where the gram functions as a
present perfect or a past perfect (pluperfect), respectively. The nuance of
anteriority may also apply to the future. In such instances, PFV approximates
the category of a future perfect. Nevertheless, this usage is relatively limited in
Arusa, as the future perfect use of PFV is only found in reported speech. In such
cases, PFV introduces an anterior event from a future temporal perspective or
mental space (13. a). In other uses that are typical of future perfects cross-
linguistically, IPFV (13. b) is usually employed, being sometimes additionally
headed by the relativizer n- (13. c).
(13) a. e-el enkaji taisere,
3-paint house tomorrow
He is going to paint the house tomorrow.
a-yelo ncere ajo aa-joki taisere e-ela enkaji
1SG-know for.sure that 3>1-tell tomorrow 3-paint.PFV house
Asian and African Studies, Volume 26, Number 1, 2017
84
I am sure he will tell me tomorrow that he has painted the whole
house.
b. te-ne-idip ne-lotu
if-REL-finish REL-come
As soon as he has finished, he will come.
c. a-idip ena itu i-lo
1SG-finish this NEG 2SG-go.IPFV
I will have this finished before you go.
While with most verbs, the PFV form typically conveys dynamic, perfectal
or past, senses (the latter ones being perfective or, less commonly, non-
perfective), with various static verbs, the meaning of PFV is present (14. a-c).
In this usage, the PFV form of a static verb seems to be synonymous to IPFV of
the same root. For instance, the sentence with a PFV verb in (14. a) is fully
equivalent to the sentence with an IPFV verb in (14. d). There are at least some
90 verbs that allow for this type of use. This constitutes circa 20% of the total
number of basic roots in Arusa (i.e. roots that are not extended by derivational
suffixes).35 Even though relatively numerous, the class of verbs whose PFV
form functions as a present seems to be close. If a new verb is coined in Arusa,
or if it is imported from another language, it adjusts to the more common
pattern exhibiting dynamic senses (perfect, perfective or past) in PFV. As a
result, the present meaning of the PFV gram can be regarded as non-productive.
(14) a. a-ta-naur-e
1SG-PFV-be.tired-PFV
I am tired.
b. a-ta-rapoʃ-e
1SG-PFV-be.satisfied-PFV
I am satisfied.
c. a-ta-gor-e
1SG-PFV-be.angry-PFV
I am angry.
d. a-naur-a
1SG-be.tired-IPFV
I am tired.
35 Further examples are: idimu ‘be able’, iroʃi ‘be heavy’, moruo ‘be old (of animate
beings)’, boita ‘be together’, seja ‘be sick’, borr ‘be polite’, iborr ‘be white’, to ‘be
try’, eron ‘be thirsty’. Interestingly, the root moi ‘be sick’ cannot be employed in this
manner, for details see KARANI, M. The Syntactic Categories and Argument Structure
in the Verbal Complex.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
85
In order to express the idea of a state in the past, the IPFV form of given
stative verb is used, e.g. anaura (nakata) ‘I was tired (that time)’. Since the
same form typically implies present reference time (anaura usually means ‘I
am tired’), overt past-time expressions must be used (15. a). If an explicit past
temporal adverb or locution is absent, the past reference should be recoverable
from the general context. This behavior of stative verbs is similar to the use of
IPFV of all the other verbs, as such forms may refer both to the present and past
time frame (15. b-c). In case of dynamic verbs, IPFV typically expresses
habitual or progressive nuances (15. d).
(15) a. a-naur-a ŋole
1SG-be.tired-IPFV yesterday
I was tired yesterday.
b. a-yielo (tenakata)
1SG-know.IPFV (now)
I know it (now).
c. a-yielo ŋole
1SG-know.IPFV yesterday
I knew it yesterday.
d. a-itibir-iʃo ŋole
1SG-work-PROG yesterday
I was working yesterday.
Even though the PFV gram of various static verbs most commonly conveys
the idea of a present state, the same form can also be used with a dynamic
ingressive reading. As explained, this dynamic interpretation is a rule in case of
all the other verbs, which function as a present perfect or a past in PFV. In this
usage, PFV of a static predicate communicates the action of getting into the
state rather than the state itself. This dynamic use is limited to a recent past
sphere, being ungrammatical if the temporal reference exceeds hesternal past
and, in particular, if it is distant or remote (compare a similar time span of the
analytical construction with the verb idip ‘to finish’ discussed previously):
(16) kaɲoo pe-itu intibirr
why SUB-2.NEG 2.do.IPFV
Why didn’t you do it?
a-ta-naur-e ŋole
1SG-PFV-be.tired-PFV yesterday
I got tired yesterday.
Asian and African Studies, Volume 26, Number 1, 2017
86
The PFV gram can also be used in unreal counterfactual conditional periods,
both in protases and apodoses. In such instances, PFV introduces events or
actions that are contrary to the situation that is (or was) actually observed in the
real world and that, moreover, cannot be altered. In this usage, PFV
approximates the category of past irrealis:
(17) te-ne-mutu i-ta-raŋ-a ŋole a-nata i-ta-yelo
osingolio
If-REL-have 2-PFV-sing-PFV yesterday 1SG.would 2SG-
PFV-know song
If you had sung yesterday, you would have known the song.
In contrast, if the conditional period is aimed to convey the meaning of real
factuality (which corresponds to present realis), the IPFV form is regularly used.
In protases, IPFV is typically linked to the conditional conjunction by means of
the relativizer n-. In this function, the use of PFV is ungrammatical:
(18) te-ni-tum impesai, ni-kiwuo osokoni taisere
if-REL-get money REL-go shop tomorrow
If you get the money, we will go to the shop tomorrow.
In some languages, forms that are compatible with various perfectal
functions may also express a resultative proper sense from which they
historically derive. The resultative proper sense is an internally complex value.
It communicates the idea that the current state has resulted from a prior action,
both elements (i.e. the prior action and the posterior state) being equally
relevant. Forms that express the resultative proper meaning (so-called
resultative proper grams) are typically intransitive and/or de-transitive,
exhibiting cross-linguistic convergence with passives (e.g. it is written in
English). These two properties distinguish a resultative proper sense/gram from
the category of a present perfect resultative, which is dynamic and transitive.
That is, perfects focus on the dynamic action (the resulting state, even though
inferable, being secondary) and exhibit the same valency pattern as the
underlying root.36
36 NEDJALKOV, V., JAXONTOV, S. The Typology of Resultative Constructions. In
NEDJALKOV, V. (ed.). Typology of Resultative Constructions, pp. 362; MASLOV, J.
Resultative, Perfect and Aspect. In NEDJALKOV, V. (ed.). Typology of Resultative
Constructions, pp. 96113; NEDJALKOV, V. Resultative Constructions. In
HASPELMATH, M. et al. (eds.) Language Typology and Language Universals: An
International Handbook, Vol. 2, pp. 928940.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
87
In Arusha, the use of PFV with a resultative proper sense is ungrammatical.
The resultative proper sense in a present time frame is regularly expressed by
the passive IPFV gram, or, more accurately, the impersonal IPFV form (19. a).
The “bare” PFV form of a dynamic (transitive) verb never exhibits a resultative
proper value. If the impersonal form is inflected in PFV, it expresses the TAM
nuances that are fully consistent with the uses of PFV discussed above. That is,
the construction is used as a passive perfect or past (perfective or non-
perfective; 19. b).
(19) a. e-duŋ-o olcani. E-perr te enkop
3-cut-PFV tree 3-lay.IPFV on ground
The tree is cut. It lies on the ground.
b. e-tu-duŋ-o-ki olcani
3-PFV-cut-PFV-IMP tree
The tree was / has been cut.
As is evident from the previous discussion, the PFV gram exhibits a wide
range of semantic potential. All these senses are, however, restricted to positive
(affirmative) contexts. In contrast, the PFV form fails to be employed in the
negative. The negative equivalent of PFV is expressed by a complex
construction built of the element (e)itu and the lexical verb inflected in IPFV.
This construction applies to all the semantic variants of PFV discussed
previously in this section, be it perfect (present perfect, past perfect and future
perfect; 20. a), past (perfective and non-perfective; 20. b) or stative present (20.
c). This same holds true for modal (counterfactual) uses.
(20) a. itu aikata a-ɲa olmaembei
NEG never 1SG-eat.IPFV mango
I have never eaten a mango.
b. itu a-lot-ie olbasi ŋole
NEG 1SG-go.IPFV-INST bus yesterday
I didn’t travel by bus yesterday.
c. itu a-naura
Neg 1SG-be.tired.IPFV
I am not tired.
This negative construction is also employed in cases where one wants to
indicate that an action occurred before another event, either in present, future
(21. a) or past (21. b):
Asian and African Studies, Volume 26, Number 1, 2017
88
(21) a. wou eton itu a-lo37
come before NEG 1SG-go.IPFV
Come before I go (come, I haven’t gone yet).
b. e-ewu-o yeyio itu a-lo
3-come-PFV my.mother NEG 1SG-go.IPFV
My mother had come, before I left.
In the negative construction discussed above, the element (e)itu is a PFV
verbal form.38 Its exact origin is, however, unknown.39 In Arusa, this verbal
element is still inflected appearing as itu in the 1st and 2nd person (singular and
plural), and eitu in the 3rd person (singular and plural).
4. Discussion
The qualitative evidence provided in the previous section demonstrates that
PFV is compatible with the following semantic domains: present perfect,
definite past, pluperfect and future perfect. In the domain of a present perfect,
PFV is compatible with resultative and experiential nuances. In the domain of a
definite past, it communicates any degree of remoteness from the speaker’s now,
and introduces both perfective or non-perfective (durative and repetitive-
habitual) events and/or activities. With static roots, PFV expresses present states.
However, static roots may also be used dynamically in the sense of a present
perfect or recent (hodiernal/hesternal) past. Lastly, PFV appears in conditional
protases and apodoses where it communicates the idea of unreal counter
factuality.
Having identified the components of the semantic potential of PFV, the
question arises as for how to organize it into a map. This question relates to the
following query: what typologically viable grammaticalization path may have
driven the PFV form to acquire the array of senses observed in Arusa? As will
be evident from the subsequent discussion, the conceptual and diachronic
coherence of PFV can be achieved by matching the semantic potential of this
form with the resultative path and paths that are related to it.
A resultative path is a complex evolutionary scenario. It models the lifecycle
of constructions that emerge as resultative proper grams, showing the order in
37 TUCKER, A. N., MPAAYEI, J. T. O. A Grammar of Maasai with Vocabulary, p. 102.
38 KOOPMAN, H. On the Homophony of Past Tense and Imperatives in Kisongo
Maasai. In UCLA Working Papers in Linguistics, 2001, Vol. 6, p. 4.
39 TUCKER, A. N., MPAAYEI, J. T. O. A Grammar of Maasai with Vocabulary, p. 101.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
89
which such grams acquire new senses.40 The resultative path consists of two
sub-paths: an anterior path (which leads towards a past sense/gram) and a
simultaneous path (which leads towards a present sense/gram).41
By travelling along the anterior path, grams whose original sense is a
resultative proper present acquire senses of a present perfect and subsequently
of a definite past. The acquisition of the perfectal values typically begins with
an inclusive sense and ends with an experiential sense, passing through the
sense of a resultative present perfect.42 When used as a definite past, the gram
commonly expands its temporal distance from less distant (e.g. immediate >
hodiernal > hesternal > recent) to more distant (general > remote).43 In some
languages, an aspectual development within the definite-past domain is
observed, the form being first compatible with perfective and next with non-
perfective senses: durative, iterative-habitual, and progressive.44
40 Sometimes completives are considered as a separated sub-type, see BYBEE, J.,
PERKINS, R., PAGLIUCA, W. The Evolution of Grammar. Even though the meaning of
lexical sources of completives is distinct from the meaning of resultative proper
constructions, their acquisition of perfectal, perfective and past senses is analogous to that
exhibited by resultative proper grams (on completives, see further below in this section).
41 ANDRASON, A. From Resultatives to Present Tenses – Simultaneous Path of
Resultative Constructions. In Italian Journal of Linguistics, 2014, Vol. 26, No. 1, pp. 1
58; ANDRASON, A. A Complex System of Complex Predicates: Tense, Taxis, Aspect
and Mood in Basse Mandinka from a Grammaticalization and Cognitive Perspective;
ANDRASON, A. From Vectors to Waves and Streams: An Alternative Approach to
Semantic Maps. In Stellenbosch Papers in Linguistics, 2016, Vol. 45, pp. 129; BYBEE,
J., PERKINS, R., PAGLIUCA, W. The Evolution of Grammar.
42 ANDRASON, A. From Resultatives to Present Tenses – Simultaneous Path of
Resultative Constructions. In Italian Journal of Linguistics, 2014, Vol. 26, No. 1, pp. 1
58; ANDRASON, A. A Complex System of Complex Predicates: Tense, Taxis, Aspect
and Mood in Basse Mandinka from a Grammaticalization and Cognitive Perspective.
43 BYBEE, J., PERKINS, R., PAGLIUCA, W. The Evolution of Grammar.
44 NEDJALKOV, V., JAXONTOV, S. The Typology of Resultative Constructions. In
NEDJALKOV, V. (ed.). Typology of Resultative Constructions, pp. 363; BYBEE, J.,
PERKINS, R., PAGLIUCA, W. The Evolution of Grammar, pp. 5557, 98, 104105;
SQUARTINI, M., BERTINETTO, P. M. The Simple and Compound Past in Romance
Languages. In DAHL, Ö. (ed.). Tense and Aspect in the Languages of Europe, pp. 406
407, 414417, 422; DAHL, Ö. The Tense and Aspect Systems of European Languages
in a Typological Perspective. In DAHL, Ö. (ed.). Tense and Aspect in the Languages of
Europe, p. 15; NEDJALKOV, V. Resultative Constructions. In HASPELMATH, M. et
al. (eds.) Language Typology and Language Universals: An International Handbook,
Vol. 2, pp. 928940; HEINE, B., KUTEVA, T. The Genesis of Grammar: A
Reconstruction, p. 151; ANDRASON, A. A Complex System of Complex Predicates:
Tense, Taxis, Aspect and Mood in Basse Mandinka from a Grammaticalization and
Cognitive Perspective.
Asian and African Studies, Volume 26, Number 1, 2017
90
In contrast, by travelling along the simultaneous path, a gram used as a
resultative proper present acquires the sense of a stative present and, later, a
non-stative present.45
The anterior path principally operates for dynamic verbs, while the
simultaneous path tends to be travelled by static verbs and by verbs that easily
lend themselves for stative interpretations (e. g. sensory verbs). However, the
anterior path may gradually attract all predicates, be they dynamic or static.46
This is related to a greater cross-linguistic pervasiveness of the anterior path if
compared to the simultaneous path.
Resultative proper grams can also start their grammatical evolution within a
past and future time frame. This gives rise to past perfect (pluperfects) and past
stative values, on the one hand, and future perfect and future stative values, on
the other. Subsequently, grams that express these two senses may lose their
perfectal and stative nuances and become acceptable in the function of a remote
past and simple future, respectively.47
45 MASLOV, J. Resultative, Perfect and Aspect. In NEDJALKOV, V. (ed.). Typology
of Resultative Constructions, pp. 70–71; BYBEE, J., PERKINS, R., PAGLIUCA, W.
The Evolution of Grammar, pp. 74–78; DRINKA, B. The Evolution of Grammar:
Evidence from Indo-European Perfects. In SCHMID, M., AUSTIN, J., STEIN, D. (eds.).
Historical Linguistics 1997, p. 120; ANDRASON, A. From Resultatives to Present
Tenses Simultaneous Path of Resultative Constructions. In Italian Journal of
Linguistics, 2014, Vol. 26, No. 1, pp. 158; ANDRASON, A. A Complex System of
Complex Predicates: Tense, Taxis, Aspect and Mood in Basse Mandinka from a
Grammaticalization and Cognitive Perspective.
46 ANDRASON, A. From Resultatives to Present Tenses – Simultaneous Path of
Resultative Constructions. In Italian Journal of Linguistics, 2014, Vol. 26, No. 1, pp. 1
58; ANDRASON, A. A Complex System of Complex Predicates: Tense, Taxis, Aspect
and Mood in Basse Mandinka from a Grammaticalization and Cognitive Perspective.
The evolution of one category in both directions can be illustrated by a great number of
languages see, ANDRASON, A. El sistema verbal hebreo en su contexto semítico: una
visión dinámica [The Hebrew Verbal System in its Semitic Context: A Dynamic
Perspective]; ANDRASON, A. From Resultatives to Present Tenses – Simultaneous
Path of Resultative Constructions. In Italian Journal of Linguistics, 2014, Vol. 26, No.
1, pp. 158; ANDRASON, A. A Complex System of Complex Predicates: Tense, Taxis,
Aspect and Mood in Basse Mandinka from a Grammaticalization and Cognitive
Perspective; ANDRASON, A., LOCATELL, C. The Perfect Wave. One of the most
evident case is provided by Germanic languages, where the original “perfect” or
“resultative” category travelled the resultative path in both direction, i.e. toward the past
tense (see the Preterite ran) and the present tense (see the class of Preterite-Present
verbs such as can; compare footnote 24).
47 ANDRASON, A. El sistema verbal hebreo en su contexto semítico: una visión
dinámica [The Hebrew Verbal System in its Semitic Context: A Dynamic Perspective];
ANDRASON, A. From Resultatives to Present Tenses Simultaneous Path of
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
91
Additionally, original resultative grams which subsequently acquire
perfectal, perfective and/or past senses may develop certain modal values.
This typically occurs due to the process of modal contamination. In course of
this process, because of being persistently used in clearly modal contexts,
indicative constructions adopt modal meanings characteristic of such
environments as their own. 48 Modal contamination constitutes a type of
‘conventionalization of implicature’,49 ‘context-induced reinterpretation’,50 or
‘semantization’. 51 One of the most prototypical environments of modal
contamination are conditional protases and apodoses.52
The semantic potential exhibited by the PFV form matches the different
stages of the resultative path to a great extent. The senses of a present perfect,
perfective past and non-perfective past can be accommodated by the anterior
path. The present stative sense can be connected to those values by means of the
simultaneous path. The senses of pluperfect and future perfect are also fully
consistent with this mapping. Their position is analogous, the path (and,
possibly, the original resultative proper input) being located in a past or a future
time frame. The values of unreal counterfactuality may be linked to the
indicative semantic sphere of PFV by means of a modal contamination path that
has arisen in an overt modal environment of conditional periods.
As a result, by being organized into a vectored map of related senses, the
qualitative semantic potential of the PFV gram can be grasped in its totality. As
explained, the relation of the components of the map is both conceptual and
diachronic. The following model illustrates this profound adequacy between the
semantics of PFV and the stages of the resultative path:
Resultative Constructions. In Italian Journal of Linguistics, 2014, Vol. 26, No. 1, pp. 1
58; ANDRASON, A. A Complex System of Complex Predicates: Tense, Taxis, Aspect
and Mood in Basse Mandinka from a Grammaticalization and Cognitive Perspective.
48 ANDRASON, A. El sistema verbal hebreo en su contexto semítico: una visión
dinámica [The Hebrew Verbal System in its Semitic Context: A Dynamic Perspective];
ANDRASON, A. A Complex System of Complex Predicates: Tense, Taxis, Aspect and
Mood in Basse Mandinka from a Grammaticalization and Cognitive Perspective.
49 DAHL, Ö. Tense and Aspect Systems; BYBEE, J., PERKINS, R., PAGLIUCA, W.
The Evolution of Grammar, pp. 2526, 296.
50 HEINE, B., CLAUDI, U., HÜNNEMEYER, F. Grammaticalization. A Conceptual
Framework, pp. 7172.
51 HOPPER, P., TRAUGOTT, E. Grammaticalization, p. 82.
52 BYBEE, J., PERKINS, R., PAGLIUCA, W. The Evolution of Grammar.
Asian and African Studies, Volume 26, Number 1, 2017
92
Figure 1: Map of PFV as a vector53
Even though PFV is compatible with all the senses mentioned above and its
qualitative map spans a large portion of the resultative path, the relevance of
these senses and their contribution to the gram’s semantic potential are not
equal. Some senses are prototypical, while others are semi-prototypical or even
non-prototypical. The most prototypical sense is a perfective past and a present
perfect. However, one should bear in mind that, within the present-perfect
domain, PFV fails to be compatible with the idea of an inclusive present perfect.
This suggests a slightly lower degree of prototypicality of the domain of perfect
in comparison to the value of a perfective past. The non-perfective past uses
may be regarded as semi-prototypical. On the one hand, the PFV gram can
express duration and most cases of iterativity and habituality. On the other, it is
incompatible with the adverb ‘always, everyday’ (the most exemplary adverb of
habituality) and with progressive readings. The sense of a present stative is also
semi-prototypical – it applies to circa 20% of the roots. Its prototypicality seems
to be even lesser than that of a non-perfective past, given that the use of static
roots in PFV is not productive, and that such roots may also exhibit present
perfect and perfective past (albeit only hodiernal/hesternal) values in PFV. The
pluperfect sense is non-prototypical. The PFV gram is only commonly used in
one of its variants, i.e. when the anterior event is introduced from a past metal-
space perspective. In the other pluperfect function, it is usually reformulated
into an analytical construction built around the PFV form of the auxiliary verb
‘finish’ (see further below in this section). The future perfect exhibits a similar
behavior, being fully restricted to a future mental-space perspective of the
reported speech.
If this non-prototypicality of the past perfect and future perfect senses is
combined with certain limitations on the present perfect uses (see above in this
paragraph), the domain of taxis (i.e. all the perfectal senses in the three temporal
spheres) can be viewed as less prototypical than the perfective past. This, in
turn, suggests a semi-advancement of PFV on its grammaticalization path and
53 The shaded grey areas indicate senses with which the PFV form is currently
compatible. The black dot makes reference to the conceptual and diachronic source of
the path.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
93
its similarity with the category of “old perfects” in Bybee, Perkins and
Pagliuca’s terminology54 i.e. (perfective) past grams with common perfectal
uses. In other words, if PFV was a canonical perfect, one would expect it to
appear with no restrictions with a past and future temporal reference (i.e. as a
pluperfect and future perfect), not only with a present reference (as a present
perfect). However, as the gram is profoundly associated with the function of a
definite past, a new, more explicit construction has been derived in order to
introduce anterior actions in the past. This construction explores, itself, a
common morpho-syntactic pattern of completives, which are the other type of
grams that “feed” the resultative path, especially its anterior subtype.55 It is built
around the verb ‘finish’, the most frequent lexical source of completives cross-
linguistically.56 As explained in section 3, in Arusa, this analytical gram is not
restricted to pluperfect uses but may also be used as a present perfect and
immediate, hodiernal and hesternal past. Overall, it arguably constitutes a new
(second) wave of resultative-path grams in Arusa. Having been derived after the
time where the PFV gram was coined, it is a “younger” construction. Therefore,
it has travelled the path to a lesser extent and exhibits a less advanced profile
than PFV, being prototypical in perfectal (taxis) senses.
The semi-advancement of PFV is also consistent with the previously
explained split between static and dynamic verbs, which is synchronically
typical of grams analyzed in various languages as perfects and/or perfectives.57
The modal unreal counterfactual uses are overall non-prototypical, being
limited to one modally marked environment – conditional periods. However, in
this environment, the PFV form is the most common gram that is used to
communicate unreal counterfactual nuances. In general, the exclusive
compatibility with the modal value of unreal counterfactuality and hence the
incompatibility with real factual ideas is coherent with the semi-advanced
grammaticalization stage of PFV and its primary association with the idea of
perfective past and definite past.58
54 BYBEE, J., PERKINS, R., PAGLIUCA, W. The Evolution of Grammar.
55 BYBEE, J., PERKINS, R., PAGLIUCA, W. The Evolution of Grammar, p. 105.
56 BYBEE, J., PERKINS, R., PAGLIUCA, W. The Evolution of Grammar, p. 5861
57 BYBEE, J., PERKINS, R., PAGLIUCA, W. The Evolution of Grammar; see also
BOTNE, R. On the Notion ‘Inchoative Verb’ in Kinyarwanda. In JOUANNET, F. (ed.).
Le kinyarwanda, études linguistiques; BOTNE, R. On the Nature of Tense and Aspect.
Studies in the Semantics of Temporal Reference in English and Kinyarwanda; BOTNE,
R., KERSHNER, T. Time, Tense and the Perfect in Zulu. In Africa und Übersee, 2000,
Vol. 83, pp. 161180; BAR-EL, L. Aspectual Distinctions in SKWXWU7MESH;
FLEISCH, A. Lucazi Grammar. A Morphosemantic Analysis.
58 Compare BYBEE, J., PERKINS, R., PAGLIUCA, W. The Evolution of Grammar;
ANDRASON, A. (2013b). Against Floccinaucinihilipilification of the Counterfactual
Asian and African Studies, Volume 26, Number 1, 2017
94
Given the range of senses and their degree of prototypicality, the qualitative-
quantitative map of the PFV form adopts the following wave shape:59
Figure 2: Map of PFV as a wave
So far, the definition of the PFV gram in terms of a semi-advanced
resultative-path gram has been proposed on a synchronic and typological
ground. That is, the variation of senses exhibited currently by PFV has been
matched with a typologically pervasive grammaticalization path and understood
as a language-specific manifestation of that path. The cross-linguistic
“universality” of the resultative path enabled us to postulate the mapping even
though the actual evolution of the form has not been studied.60
Presently, scholars do not have access to direct diachronic evidence that
could show how the PFV form has emerged and developed in Arusa and Maasai
(e.g. texts attested as different historical periods). Nevertheless, the proposed
Tense of the BH Suffix Conjugation – or an Explanation of Why the “Indicative” Qatal
Expresses Conditions, Hypotheses and Wishes. In Old Testament Essays, 2013, Vol. 26,
No. 1, pp. 2056; ANDRASON, A. A Complex System of Complex Predicates: Tense,
Taxis, Aspect and Mood in Basse Mandinka from a Grammaticalization and Cognitive
Perspective.
59 In this paper, four values of prototypicality have been distinguished: prototypical (P),
semi-prototypical (SP), non-prototypical (NP) and void (V). Certainly, more fine-
grained distinctions are possible, especially if exact quantitative data are collected, i.e.
frequency. In such cases, the degree of prototypicality would range from 0% to 100%
through any level of granularity (see Andrason 2016b).
60 Compare HEINE, B. Cognitive Foundations of Grammar.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
95
mapping of PFV may be corroborated by certain structural and systemic
properties that indirectly reveal the diachronic origin of this form.
It has been suggested that in Maasai, PFV and the Imperative share, to an
extent, their origin, having derived from the same “syntactic substructure”,
some type of a participial form.61 That is, the shared properties of PFV and the
Imperative and their systematic patterning are not accidental but reveal a
possible shared “ancestor” of the two constructions. 62 These similarities
involve: (a) morphological homonymy – i.e. the use of the affix -tV- for Class I
verbs; (b) equivalent use of suppletive forms – i.e. both categories employ
identical stems in case of suppletive verb; and (c) incompatibility with negation
i.e. PFV is negated by auxiliary (e)itu and the IPFV form of a lexical verb,
while the Imperative is negated by a negative subjunctive.63
The relationship between PFV and the Imperative, and their possible
participial origin as reconstructed by Koopman64 harmonize with our mapping.
This stems from the following cross-linguistic observation: grams of the
resultative path (such as PFV) are in various languages derived from some type
of participial constructions and exhibit connection with the category of
imperative or with imperative uses.
To be exact, examples of resultative-path grams that have been derived from
participial constructions (in particular passive, patientive, resultative or perfect
participles) are fairly abundant cross-linguistically. Such grams are especially
common in Indo-European languages where they have arisen independently, for
instance, in Romance, Germanic, Slavonic, Greek and Indo-Iranian.65 They are
also well attested in the Semitic family, e.g. in Akkadian, Hebrew, Arabic and
61 KOOPMAN, H. On the Homophony of Past Tense and Imperatives in Kisongo
Maasai. In UCLA Working Papers in Linguistics, 2001, Vol. 6, pp. 1, 5, 6.
62 Ibid., p. 2.
63 Ibid., pp. 2, 4, 5. In Maasai, the Imperative is not the morphologically simplest (i.e.
the most impoverished) form, as is common cross-linguistically. Rather than being a
“true imperative”, the Maasai Imperative is a “dependent form”. It should be noted that
in the Imperative, contrary to other verbal forms such as PFV, the pronominal
morphemes do not precede but follow the verbal stem (KOOPMAN, H. On the
Homophony of Past Tense and Imperatives in Kisongo Maasai. In UCLA Working
Papers in Linguistics, 2001, Vol. 6, pp. 67).
64 KOOPMAN, H. On the Homophony of Past Tense and Imperatives in Kisongo
Maasai. In UCLA Working Papers in Linguistics, 2001, Vol. 6.
65 BYBEE, J., PERKINS, R., PAGLIUCA, W. The Evolution of Grammar;
ANDRASON, A. From Resultatives to Present Tenses Simultaneous Path of
Resultative Constructions. In Italian Journal of Linguistics, 2014, Vol. 26, No. 1, pp. 1
58.
Asian and African Studies, Volume 26, Number 1, 2017
96
Aramaic.66 Further cases include Basque, Tucanoan family (Tucano) and the
Dravidian family (Kui). 67 The relation between resultative-path grams and
imperatives (and the semantic domain of orders) is also relatively pervasive
cross-linguistically. That is, grams that function as indicative perfects,
perfectives and pasts either exhibit an identical or similar shape to imperative-
type constructions, or are themselves used to communicate orders and
commands. This may be illustrated by Slavonic (Polish and Russian), Semitic
(Akkadian, Hebrew, Arabic) and Mande (Mandinka) examples.68 For example,
in Polish the L and N/T past senses (napisał ‘he wrote’ and napisano ‘one
wrote’) can be used in commands: Napisano mi to teraz! ‘Write it now!’ and
Poszedł stąd! ‘Go away from here!’.69
To conclude, if the PFV gram originated in a participial input that also fed
the Imperative form, our definition of PFV as a resultative-path gram is
corroborated diachronically apart from being derived by matching the
synchronic evidence with typological tendencies.70
5. Conclusion
The present paper studied the semantics of the PFV form in Arusa, using the
model of dynamic semantic maps which emerges at the interfaces of cognitive
linguistics and grammaticalization theory. Depending on the type of data
included, the map surfaced as a vector or a wave. The former builds on
qualitative data, while the latter draws both from qualitative and quantitative
data. In this study, qualitative data corresponded to the senses that the PFV
66 ANDRASON, A. El sistema verbal hebreo en su contexto semítico: una visión
dinámica [The Hebrew Verbal System in its Semitic Context: A Dynamic Perspective].
67 BYBEE, J., PERKINS, R., PAGLIUCA, W. The Evolution of Grammar.
68 For a detailed discussion see ANDRASON, A. El sistema verbal hebreo en su
contexto semítico: una visión dinámica [The Hebrew Verbal System in its Semitic
Context: A Dynamic Perspective]; ANDRASON, A. An Optative Indicative? A Real
Factual Past? A Cognitive-Typological Approach to the Precative Qatal. In Journal of
Hebrew Scriptures, 2013, Vol. 13, No. 4, pp. 141; ANDRASON, A. A Complex
System of Complex Predicates: Tense, Taxis, Aspect and Mood in Basse Mandinka from
a Grammaticalization and Cognitive Perspective.
69 ANDRASON, A. An Optative Indicative? A Real Factual Past? A Cognitive-
Typological Approach to the Precative Qatal. In Journal of Hebrew Scriptures, 2013,
Vol. 13, No. 4, pp. 141.
70 Overall, our analysis of the PFV form and its definition as a resultative-path gram
corroborates the reconstruction proposed by Koopman, see KOOPMAN, H. On the
Homophony of Past Tense and Imperatives in Kisongo Maasai. In UCLA Working
Papers in Linguistics, 2001, Vol. 6.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
97
form conveys, whereas quantitative data referred to the degree of their
prototypicality.
The analysis demonstrates that PFV can be defined as a broad resultative-
path gram. PFV spans large sections of both the anterior and the simultaneous
path. However, the input sense of the two paths (i.e. a resultative proper
present) and their most advanced sections (i.e. a non-stative present and a
progressive past) do not belong to the semantic potential of the gram. The
former value (resultative proper) has likely been lost, whereas the latter values
(non-stative present and progressive past) have not been acquired yet. If the
information related to prototypicality is included, the map takes on the shape of
a wave with the prototypicality peaks located in the area of present perfect and,
especially, perfective past. The senses of a non-perfective past and a stative
present are less prototypical – therefore, at these points, the wave is less raised.
Other values are non-prototypical, contributing minimally to the gram’s
semantics. Although the vectored and waved maps were principally posited on
the synchronic and typological grounds, they were also corroborated by certain
structural and systemic properties. These properties relate to a participial origin
of the PFV form, and its connection to the Imperative.
The results of this article may arguably contribute to the aspect-tense debate,
pervasive in Maasai scholarship. This debate concerns the nature of PFV and
the entire verbal system i.e. its aspectual, temporal or taxis (i.e. perfectal)
foundation. In our view, the one form one meaning approach” should be
replaced by a more nuanced and more empirically-grounded model of maps.
That is, as any verbal gram, the PFV form is polysemous and draws from
various domains. It is compatible with taxis, aspect and tense domains.
However, the compatibility with each domain is found in different context and
amounts to a different extent.
Our study has not answered all the questions related to the semantics of the
PFV gram. In particular, the quantitative information specifying the degree of
prototypicality of senses was coarse-grained. It was also relatively approximate
and in some cases based on the intuition of native speakers. Certainly more,
purely quantitative, frequency-oriented studies of large and diversified corpora
are needed. Hopefully, the efforts of the two authors to stimulate literary
production in Arusa will yield texts suitable for such quantitative analyses in the
future.
Abbreviations
DAT dative applicative;
IPFV – imperfective;
Asian and African Studies, Volume 26, Number 1, 2017
98
IMP – impersonal;
PFV – perfective;
PL – plural;
PROG – progressive;
SG singular;
SUB – subjunctive.
REFERENCES
ANDRASON, Alexander. (2013a). El sistema verbal hebreo en su contexto
semítico: una visión dinámica. [The Hebrew Verbal System in its Semitic
Context: A Dynamic Perspective]. Estella: Verbo divino, 2013.
ANDRASON, Alexander. (2013b). Against Floccinaucinihilipilification of the
Counterfactual Tense of the BH Suffix Conjugation or an Explanation of
Why the “Indicative” Qatal Expresses Conditions, Hypotheses and Wishes.
In Old Testament Essays, 2013, Vol. 26, No. 1, pp. 2056.
ANDRASON, Alexander. (2013c). An Optative Indicative? A Real Factual
Past? A Cognitive-Typological Approach to the Precative Qatal. In Journal
of Hebrew Scriptures, 2013, Vol. 13, No. 4, pp. 141.
ANDRASON, Alexander. From Resultatives to Present Tenses Simultaneous
Path of Resultative Constructions. In Italian Journal of Linguistics, 2014,
Vol. 26, No. 1, pp. 158.
ANDRASON, Alexander. (2016a). A Complex System of Complex Predicates:
Tense, Taxis, Aspect and Mood in Basse Mandinka from a
Grammaticalization and Cognitive Perspective. Ph.D. dissertation.
Stellenbosch: Stellenbosch University, 2016.
ANDRASON, Alexander. (2016b). From Vectors to Waves and Streams: An
Alternative Approach to Semantic Maps. In Stellenbosch Papers in
Linguistics, 2016, Vol. 45, pp. 129.
ANDRASON, Alexander, LOCATELL, Christian. The Perfect Wave. In
Biblical and Ancient Greek Linguistics, 2016, Vol. 5, pp. 7122.
BAR-EL, Leora. Aspectual Distinctions in SKWXWU7MESH. Ph.D.
dissertation. Vancouver: The University of British Columbia, 2005.
BOTNE, Robert. On the Notion ‘Inchoative Verb’ in Kinyarwanda. In
JOUANNET, Francis (ed.). Le kinyarwanda, études linguistiques. Paris:
SELAF, 1983, pp. 149180.
BOTNE, Robert. On the Nature of Tense and Aspect. Studies in the Semantics
of Temporal Reference in English and Kinyarwanda. Ann Arbor: UMI, 1987.
BOTNE, Robert, KERSHNER, Tiffany. Time, Tense and the Perfect in Zulu. In
Africa und Übersee, 2000, Vol. 83, pp. 16180.
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
99
BYBEE, Joan. Language, Usage and Cognition. Cambridge: Cambridge
University Press, 2010.
BYBEE, Joan, PERKINS, Revere, PAGLIUCA, William. The Evolution of
Grammar. Chicago: The University of Chicago Press, 1994.
COMRIE, Bernard. Aspect: An Introduction to the Study of Verbal Aspect and
Related Problems. Cambridge: Cambridge University Press, 1976.
COUSSÉ, Evie. Lexical Expansion in the HAVE and BE Perfect in Dutch A
Constructionist Prototype Account. In Diachronica, 2014, Vol. 31, pp. 159
191.
COUSSÉ, Evie. Grammaticalization, Host-class Expansion and Category
Change. In NORDE, Muriel, VAN GOETHEM, Kristel, COUSSÉ, Evie,
VANDERBAUWHEDE, Gudrun (eds.). Category Change from a
Construction Grammar Perspective. Amsterdam: Benjamins, forthcoming.
DAHL, Östen. Tense and Aspect Systems. Oxford: Blackwell, 1985.
DAHL, Östen. The Tense and Aspect Systems of European Languages in a
Typological Perspective. In DAHL, Östen (ed.). Tense and Aspect in the
Languages of Europe. Berlin: Mouton de Gruyter, 2000, pp. 3–25.
DEHAAN, Ferdinand. Typology of Tense, Aspect, and Modality Systems. In
SONG, J. J. (ed.). The Oxford Handbook of Linguistic Typology. Oxford:
Oxford University Press, 2011, pp. 445–464.
DRINKA, Bridget. The Evolution of Grammar: Evidence from Indo-European
Perfects. In SCHMID, Monika, AUSTIN, Jennifer, STEIN, Dieter (eds.).
Historical Linguistics 1997. Amsterdam: John Benjamins, 1998, pp. 117–133.
FLEISCH, Axel. Lucazi Grammar. A Morphosemantic Analysis. Cologne:
Köppe, 2000.
HAMAYA, Mitsuyo. Vowel Harmony in Maasai. In Linguistics, 1997, Vol. 607,
pp. 130.
HASPELMATH, Martin. The Geometry of Grammatical Meaning Semantic
Maps and Cross-linguistic Comparison. In THOMASELLO, Michael (ed.).
The New Psychology of Language. Mahwah: Lawrence Erlbaum Associates,
2003, pp. 211242.
HASPELMATH, Martin. Coordinating Constructions: An Overview. In
HASPELMATH, Martin (ed.). Coordinating Constructions. Amsterdam:
John Benjamins, 2004, pp. 140.
HEINE, Bernd. Cognitive Foundations of Grammar. Oxford: Oxford University
Press, 1997.
HEINE, Bernd, CLAUDI, Ulrike, HÜNNEMEYER, Friederike. Grammatica-
lization. A Conceptual Framework. Chicago: University of Chicago Press,
1991.
HEINE, Bernd, KUTEVA, Tania. The Genesis of Grammar: A Reconstruction.
Oxford: Oxford University Press, 2007.
Asian and African Studies, Volume 26, Number 1, 2017
100
HOLLIS, A. C. The Maasai. Their Language and Folklore. Oxford: Clarendon
Press, 1905.
HOPPER, Paul, TRAUGOTT, Elisabeth. Grammaticalization. Cambridge:
Cambridge University Press, 2003.
JANDA, Laura. Cognitive Linguistics in the Year 2015. In Cognitive Semantics,
2015, Vol. 1, pp. 131154.
KARANI, Michael. The Arusa Verb System. M. A. dissertation. Dar se Salam:
University of Dar es Salaam, 2013.
KARANI, Michael. The Syntactic Categories and Argument Structure in the
Verbal Complex. Ph.D. dissertation. Stellenbosch: Stellenbosch University,
forthcoming.
KONIG, Christa. Aspekt im Maa. Cologne: University of Köln, 1993.
KOOPMAN, Hilda. On the Homophony of Past Tense and Imperatives in
Kisongo Maasai. In UCLA Working Papers in Linguistics, 2001, Vol. 6, pp.
1–13.
LEVERGOOD, Barbara Jo. Topics in Arusa Phonology and Morphology. Ph.D.
dissertation. Austin: University of Texas, 1987.
MCCAWLEY, Jim. Tense and Time Reference in English. In FILLMORE,
Charles J., LANGENDOEN, D. Terence (eds.). Studies in Linguistics and
Semantics. New York: Holt, Rinehart and Winston, 1971, pp. 96–113.
MASLOV, Jurji. Resultative, Perfect and Aspect. In NEDJALKOV, Vladimir
(ed.). Typology of Resultative Constructions. Amsterdam: John Benjamins,
1988, pp. 6385.
NEDJALKOV, Vladimir. Resultative Constructions. In HASPELMATH,
Martin et al. (eds.). Language Typology and Language Universals: An
International Handbook. Vol. 2. Berlin: De Gruyter, 2001, pp. 928–940.
NEDJALKOV, Vladimir, JAXONTOV, Sergej. The Typology of Resultative
Constructions. In NEDJALKOV, Vladimir (ed.). Typology of Resultative
Constructions. Amsterdam: John Benjamins, 1988, pp. 3–62.
NURSE, Derek. Tense and Aspect in Bantu. Oxford: Oxford University Press,
2008.
PAYNE, Doris, HAMAYA, Mitsuyo, JACOBS, Peter. Active, Passive, and
Inverse in Maasai. In GIVÓN, Talmy (ed.). Voice and Inversion. Amsterdam:
John Benjamins, 1994, 283315.
PAYNE, Doris, OLE-KOTIKASH, Leonard, OLE-LEKUTIT, Keswe Mapena.
A Frame Semantics Approach to Lexemic Structure: Uncovering the Truth
about Maa a-síp. In Journal of African Languages and Linguistics, 2001, Vol.
22, pp. 123.
PAYNE, Doris, OLSEN, Derek. Maa (Maasai) Nominalization: Animacy,
Agentivity and Instrument. In MATONDO, Masangu et al. (ed.). Selected
The Perfective Form in Arusa – a Cognitive-Grammaticalization Model
101
Proceedings of the 38th Annual Conference on African Linguistics.
Somerville: Cascadilla Proceedings Project, 2009, pp. 151–165.
SQUARTINI, Mario, BERTINETTO, Pier Marco. The Simple and Compound
Past in Romance Languages. In DAHL, Östen (ed.). Tense and Aspect in the
Languages of Europe. Berlin: De Gruyter Mouton, 2000, pp. 385–402.
TUCKER, Archibald N., MPAAYEI, J. Tompo Ole. A Grammar of Maasai
with Vocabulary. London: Longmans, Greek and Co., 1955.
... In addition, a group of morphologically irregular verbs expresses the perfective aspect by suppletion, for example, ee 'he will die' etwa 'he died' (see Koopman 2001 for related discussion on the Kisongo dialect). Andrason and Karani (2017b) present pertinent discussion on aspect in Arusa. ...
... Thus, when -i is suffixed to the verb, as in (11d), it yields a passive reading 'the house has been plastered. For discussion of the perfective aspect, see Andrason and Karani (2017b). Evidence from the Parakuyo data suggests that the function of the impersonal morpheme -i is not only attested in the perfective aspect but also in the imperfective aspect. ...
... However, in some contexts, the imperfective occurs referring to the past if the event denoted by the verb was happening at the time that another event was happening (see Andrason and Karani 2017b for discussion on the perfective aspect in Arusa, a dialect of Maa). The following examples in (34) illustrate the interpretations of the three viewpoints that obtain in the reciprocal verb construction in the imperfective aspect. ...
... In addition, a group of morphologically irregular verbs expresses the perfective aspect by suppletion, for example, ee 'he will die' etwa 'he died' (see Koopman 2001 for related discussion on the Kisongo dialect). Andrason and Karani (2017b) present pertinent discussion on aspect in Arusa. ...
... Thus, when -i is suffixed to the verb, as in (11d), it yields a passive reading 'the house has been plastered. For discussion of the perfective aspect, see Andrason and Karani (2017b). Evidence from the Parakuyo data suggests that the function of the impersonal morpheme -i is not only attested in the perfective aspect but also in the imperfective aspect. ...
... However, in some contexts, the imperfective occurs referring to the past if the event denoted by the verb was happening at the time that another event was happening (see Andrason and Karani 2017b for discussion on the perfective aspect in Arusa, a dialect of Maa). The following examples in (34) illustrate the interpretations of the three viewpoints that obtain in the reciprocal verb construction in the imperfective aspect. ...
... Arusa is a heavily under-researched variety of Maasai. The entire collection of scholarly literature on Arusa is limited to four publications: a PhD dissertation dedicated to phonology and morpho-phonology presented by Levergood in 1987; an MA thesis dedicated to the morpho-semantics of the verbal system written by Karani in 2013; and, more recently, two papers published by the authors of the present article, of which one deals with the tense, aspect, mood (TAM) semantics of the so-called perfective form (Andrason and Karani 2017a), and the other analyzes the phenomenon of left dislocation (Andrason and Karani 2017b). ...
Article
Full-text available
The present paper analyzes the degree of the argumenthood or adjuncthood of elements licensed by the dative applicative (DA) construction in Arusa within a canonical approach to the argument-adjunct distinction. After testing DA elements for the various criteria and diagnostics associated with the typologically-driven prototype of arguments and adjuncts, the authors conclude the following: in most cases, DA elements behave as canonical arguments and are therefore located close to the argumenthood pole of the argument-adjunct continuum.
... This model has been developed by Andrason in his work on the verbal system of Mandinka (2016a, b). Subsequently, it has been applied to verbal systems of Hebrew(Andrason 2015), Greek(Andrason, Locatell 2016) and Maasai(Andrason, Karani 2017). ...
Article
Full-text available
The present paper offers an analysis of the TAM semantics of the HĨ and the HA gram(matical construction)s in Tjwao within the cognitive and grammaticalization-based model of dynamic maps and streams. The authors show that, albeit similar, the ranges of meanings of the two grams differ. The grams share the senses of experiential present perfect, definite past, stative and non-stative present. However, the senses of narrative remote past and pluperfect are typical of HĨ, while the senses of inclusive and resultative present perfect are only compatible with HA. When used as presents, HA is limited to affirmative contexts, whereas HĨ is restricted to negative contexts. The authors additionally demonstrate that the polysemy of each gram can be mapped by means of two sub-paths of the resultative path, i.e. the anterior and simultaneous paths. The two grams may therefore be located on the same stream on which HĨ occupies a more advanced position than HA, being thus a chronologically earlier construction. This grammaticalization-based entanglement of HĨ and HA is consistent with the situation found in other Khoe languages.
Thesis
Full-text available
The Turkic language Kazakh has a remarkably large set of eighteen auxiliary verbs, which express various tense-aspect-modality (TAM) values in combination with a lexical verb. This thesis presents the first multivariate analysis of auxiliary verb constructions (AVCs) to precisely define the function and distribution of every auxiliary in Modern Spoken Kazakh. Based on corpus data and first-hand fieldwork, I demonstrate that contemporary Kazakh has 28 AVCs, each with a distinctive function. I argue that some morphologically or syntactically ambiguous constructions, hitherto analyzed as AVCs, should be treated as lexical verbs in separate clauses. Previ-ously undescribed AVCs are described, including an unexpected case of stem alternation. The distribu-tional analyses demonstrate that some AVCs are sensitive to syntactic parameters, such as the independ-ence of the clause they head. Others are expressions of a purely semantic feature and thus are insensitive to syntax, such as the modal abilitative. Thus, AVCs are grouped into six classes in order to contrast distinctive distributional behavior with characteristics shared across constructions. The analysis assumes that auxiliaries are periphrastic and thus are part of the lexical verb’s paradigm. Therefore, alongside AVCs, synthetic TAM expressions are investigated and the results include precise descriptions and a novel contrastive analysis of three past tense expressions. The description is complemented by an HPSG style analysis in order to present the system using a rigorous, feature-based approach. This is the first attempt to formalize a large auxiliary system with implemented solutions that lay the grounds for future work on the diachrony of auxiliaries. I propose novel semantic features that account for distinctions including boundedness, phase specification and focus. The main contribution is a systematic, synchronic, fine-grain examination of every Kazakh aux-iliary verb, which makes this complex system available for the general linguist, as well as specialists of TAM and periphrasis.
Article
The present article expands our empirical and theoretical knowledge of conative animal calls (CACs) in the languages of the world. By drawing on canonical typology and prototype theory – and by contrasting the original evidence related to the category of CACs in Arusa Maasai with the evidence concerning CACs in other languages that is currently available in scholarship – the authors design a cross-linguistic prototype of a CAC and enumerate its 18 prototypical non-formal (semantic-pragmatic) and formal (phonetic, morphological, and syntactic) features.
Article
Full-text available
This article designs a method of improving traditional, qualitative semantic maps based on grammaticalisation paths, by including both quantitative data (frequency) and information concerning a gram’s environment (the relation to the other maps). The incorporation of qualitative evidence transforms vectored maps into waves, while the introduction of the contextual factor combines waves organised along the same grammaticalisation template into a stream. The structure of a wave delivers, in turn, the statistical prototypicality of a gram (i.e. the prototypicality that is conditioned by the gram’s own wave), whereas the structure of the stream yields product prototypicality (i.e. the prototypicality that is a combination of the gram’s wave and the other waves of the stream). It is additionally hypothesised that the product prototypicality may be an overt indicator of the psychological perception of the grams by speakers.
Article
Full-text available
Cognitive linguistics views linguistic cognition as indistinguishable from general cognition and thus seeks explanation of linguistic phenomena in terms of general cognitive strategies, such as metaphor, metonymy, and blending. Grammar and lexicon are viewed as parts of a single continuum and thus expected to be subject to the same cognitive strategies. Significant developments within cognitive linguistics in the past two decades include construction grammar and the application of quantitative methods to analyses.
Article
The present paper demonstrates that the counterfactual value displayed by the BH qatal is a rightful and logical component of the total meaning of the suffix conjugation, understood as a network of conceptually and historically connected senses. The chaining procedure built on the framework of universal paths (viz. a theory of typologically highly plausible evolutionary scenarios) enables the author to relate six specific values of the counterfactual domain (real and unreal optative, real and unreal hypothetical, as well as real and unreal conditional), and to establish their diachronic and synchronic (i.e. conceptual) arrangement. Furthermore, by employing an analogical chaining method, the entire counterfactual block is linked to the dominant indicative type of the qatal and, in particular, to its perfect, perfective and past values. The results of the article also demonstrate that - contrary to widespread opinion -the optative use of the BH qatal does not reflect a loss or omission of an original apodosis. Quite the reverse, the optative value was the initial modal meaning that the Proto-Semitic *qatal- acquired in the vicinity of the particle *law and its negative varieties: the BH optative examples are, simply then, remnants of such an ancient usage.
Article
This article determines a sequence of stages that form the 'simultaneous path', a linear progression according to which certain resultative expressions develop into present tenses. First, the author hypothesizes that this type of evolution is expected to be analogous to the orderliness of the other well-established development of resultative grams during which resultative inputs evolve into anteriors, perfectives and past grams (the anterior path) traversing three verbal domains (i.e. taxis, aspect and tense). The theorized shape of the simultaneous path (from simultaneous resultatives, through statives and towards simple presents) is subsequently corroborated by a methodology referred to as 'dynamization of typology'. It is demonstrated that meanings provided by concrete grams - successors of resultative expressions - can be matched with the three hypothesized phases of this developmental path.
Article
in this chapter instead ofabstract categories (the exception being the introductions to 1 tense and aspect). Thus, when mention is made of habitual aspect, the gram type is meant. The terminology involved with the TAM areas is notoriously confusing and unfortunately far fromconsistent so that a comparison between scholarly works is not always straightforward.One of the aimsis to clarify the major terminology used in various important studies. 0.2 Tense The semantic category of tenseis usually defined as the linguistic representation oftim e.