I begin with an overview of all of the temporal-aspectual devices that were used in the narratives across languages and ages. I first consider the overall distributions of verbal devices (Section 10.1.1) and connectives (Section 10.1.2) across languages and ages. I then examine how verbal and other types of devices are related in the narratives, showing that some conjoined uses of these two types of devices increase with age in all languages (Section 10.1.3).
Distribution of all temporal-aspectual devices Table 10.1 Verbal inflections in the Indo-European languages
4 –5 years 7 years 10 years Adults ENGLISH
Simple present .26 .32 .31 .50
Present progressive .15 .13 .04 .14
Simple past .40 .45 .57 .17
Past progressive .08 .05 .03 .03
Untensed progressive .04 .04 .04 .10
Other∗ .07 .01 .01 .06
FRENCH
Pr´esent .86 .85 .70 .73
Pass´e compos´e .11 .07 .04 .04
Pass´e simple – .01 .10 .04
Imparfait .02 .07 .10 .08
Other .01 .01 .06 .13
GERMAN
Pr¨asens .77 .75 .80 .88
Perfekt .15 .13 .16 .07
Pr¨ateritum .07 .10 .04 .04
Other .01 .03 .01 .01
∗The Other category includes: ambiguous forms and occasional future, modal conditional, or subjunctive forms in all languages; the present perfect in English; theplus-que-parfaitand participial forms in French;
and thePlusquamperfektumin German.
10.1.1 Verbal devices
10.1.1.1 Verbal morphology in the Indo-European languages
Table 10.1 displays the detailed distributions of all verbal inflections that were found in English, French, and German narratives. Separate entries show the most frequent inflections, while all other scattered uses are grouped together in an additionalothercategory (pluperfect forms, conditionals, modals, subjunctives, future forms, the present perfect, as well as some ambiguous and idiosyncratic forms). In all three languages most inflections mark either the non-past or the past. Some additional cases shown in Table 10.1 consist of untensed forms, such as some noticeable untensed progressive forms in English (shown separately) or rare untensed participial forms in French (included in theothercategory). At all ages non-past markings include mostly thenon-progressive presentin English, with some less frequentpresent progressiveforms, the Frenchpr´esentand the German Pr¨asens, both of which neutralise aspect. Most past inflections in English consist of thepreterite(non-progressive past) at all ages, with some less frequent uses of the
Time
% 100
80
60
40
20
0
4 –5 yrs 7 yrs 10 yrs Adults
English German French
Figure 10.1 Overall proportions of past inflections
progressive past, as well as rare uses of thepluperfectandpresent perfect(mostly at adult age). In French past forms include the perfective past (pass´e compos´e, pass´e simple) and the imperfective past (imparfait), with some less frequent uses of theplus-que-parfait(mostly at adult age). In German past markings include the Pr¨ateritum, thePerfekt, and thePlusquamperfektum.
A first glance at these distributions shows very different uses of past vs. non- past markings. Figure 10.1 summarises these differences by further displaying the proportions of all past inflections in each language as a function of age. It can be seen that the French and German subjects in all age groups mostly use the non-past.
In contrast, English-speaking children from four/five to ten years clearly use more past inflections than the French and German children. This difference, however, disappears at the adult age, where the non-past becomes predominant in English.
10.1.1.2 Aspectual markers in Chinese
Aspectual markers in the Chinese narratives include the following: the particlele, which marks perfectivity and/or current relevance; the particlezheand the less frequent particleneand adverbialzai4, which mark imperfective aspect.
Distribution of all temporal-aspectual devices
% 100
80
60
40
20
0
4 5 yrs 7 yrs 10 yrs Adults
NPCL LE IMP
Figure 10.2 Uses and non-uses of aspect particles in Chinese
Figure 10.2 first shows the proportions of clauses in the Chinese corpus that do vs. do not contain one of these aspectual markers. It can be seen that the majority of units contain no aspectual markers at all (NPCL), regardless of age. In addition, when markers are used, the particleleis more frequent than imperfective markers. Among the clauses that do not contain aspect markers, two main types must be distinguished:
cases with special verb or clause types, which can never take these markers (see Hickmannet al. 1994 for more details); and cases that were not constrained in any way and where some marker would be expected (at least if the sentence is considered in isolation, see below). Although both types of cases occur, unconstrained ones are by far the more frequent. This first result indicates that speakers do not systematically mark aspectual distinctions across clauses, often relying on other markings and more generally on information in the preceding discourse. I return to this point below.
As previously discussed (Chapter 3), linguistic analyses have proposed different meanings for the particlele:perfectivity, particularly when it is placed directly after the verb; and current relevance, particularly when it is placed at the end of the sentence. Further analyses therefore examined in more detail uses of this particle, differentiating the following cases: uses that are verb-final, but not sentence-final;
Time
% 100
80
60
40
20
0
4–5 yrs 7 yrs 10 yrs Adults
V/S-final V-final S-final Figure 10.3 Different uses of the aspect particlelein Chinese
uses that are sentence-final, but not verb-final; and uses that are both verb-final and sentence-final. As shown in Figure 10.3, most uses of the particlelewere formally ambiguous between its two possible readings (at least on the basis of position alone), since they were both verb-final and sentence-final. These uses are frequent at all ages, although they strongly decrease after seven years. Unambiguous perfective markers (verb-final, not sentence-final) are also frequent, but only after seven years, when they increase drastically to become as frequent as verb/sentence-final uses at ten years and adult age. Other uses presumably marking unambiguously current relevance (sentence-final, not verb-final) are rare at all ages.
Although particle use is relatively rare when it is calculated over the total number of utterances in the narratives, a more qualitative analysis shows that practically all narratives did contain some aspectual markers. This analysis examines each narra- tive in order to determine its overall density of aspectual markers by differentiating the following four types of cases: narratives without any aspect marker at all (NMK);
narratives with a majority of utterances containing such markers (MK>NMK) or with a majority of utterances containing no such marker (NMK>MK); and narra- tives with as many utterances that do and do not contain such markers (NMK=MK).
Distribution of all temporal-aspectual devices Table 10.2Uses and non-uses of Chinese aspect markers within each narrative
4 –5 years 7 years 10 years Adults
CAT STORY
NMK 1 — — —
NMK>MK 7 5 7 5
MK>NMK 2 2 3 4
NMK=MK — 3 — 1
HORSE STORY
NMK — — — —
NMK>MK 7 10 8 9
MK>NMK 2 — 1 1
NMK=MK 1 — 1 —
Table 10.2 shows the number of subjects who produced narratives of these types for each story in each age group. Only one narrative at four years contains no aspect marker at all. The most frequent type of narrative contains a majority of utterances without aspect markers, but some narratives contain a majority of utterances with such markers and a few cases contain as many utterances with as without them.
I return to these different types of narratives when I focus on the discourse contexts in which uses and non-uses of aspectual particles were embedded.
10.1.2 Connectives
The connectives that were used in the narratives include a number of devices with different formal properties: co-ordinating and subordinating con- junctions, as well as adverbial expressions. Uses of these devices were of several types, such as forms serving to link grammatically main and complement clauses (e.g. Englishthat, Germandaò, Frenchque), to express causal relations (e.g. English because, Germanweil, Frenchparce que), or to mark explicit or potential temporal- aspectual distinctions. The analysis below focuses on the devices belonging to this last subset, which fell into three main types, illustrated below:sequentialdevices, regionaldevices, and otherall-purposedevices.
(a) Sequential devices:devices that mark a relation of temporal succes- sion between events, such as the following:
English: then,and then,before,after
German: und da/dann (‘and there/then’), dann (‘then’), danach (‘after that’),nachdem(‘after’)
French: et puis(‘and then’),puis‘then’,avant/avant que(‘before’), apr`es/apr`es que(‘after’)
Time
Chinese: hou4lai2(‘afterwards’),yi3hou4(‘later’, ‘after’),ran2hou4 (‘then’)
(b) Regional devices: devices that express that one situation is in the
‘temporal region’ of another, thereby indicating an explicit or potential relation of partial or total overlap, or merely a vague relation of close temporal proximity, such as the following:
English: while,meanwhile,when,(just) as
German: in der zwischenzeit(‘in the meantime’),(gerade) als(‘just when’),wenn(‘when’)
French: pendant que(‘while’),pendant ce temps-l`a (‘in the mean- time’),quand(‘when’),(juste) au moment o`u(‘(just) at the moment where’),`a ce moment-l`a(‘at this moment’) Chinese: zhei4 shi2hou4(‘at this moment’),de shi2hou4(‘while’) (c) Other all-purpose devices: these devices mainly correspond to
the uses of all-purpose co-ordinating conjunctions, when they were used alone in the Indo-European languages (Englishand, Frenchet, Germanund).
All-purpose devices are most frequent at all ages in English (occurring in 59% to 65% of units), while they are somewhat less frequent in French and German, where they also gradually increase from four/five years (33% and 27%, respectively) to adult age (50% and 61%, respectively). In addition, as shown in Figure 10.4, regional devices increase with age, whereas sequential ones decrease in all four languages.
However, the rate of this progression varies across languages. In particular, in the three Indo-European languages regional devices mostly increase between seven and ten years (French) or between ten years and adult age (English, German). In comparison, this increase begins earlier and it is more striking in Chinese, where regional devices first increase between four and seven years and steadily continue to do so thereafter until adult age. Overall, although regional devices are infrequent at four/five years in all languages, Chinese subjects use them more frequently than other language groups thereafter, especially at seven and ten years.
10.1.3 Relation between verbal and other devices
Table 10.3 shows the proportions of the different inflections and par- ticles that were used with regional vs. sequential devices at each age in the four languages. In English the simple present, the present progressive, and the simple past are most frequent with regional devices. The past increases with these devices until ten years as to a lesser extent do the simple and progressive present at adult age. The progressive is rare with sequential devices. In French all inflection types occur with sequential devices at seven and ten years, but the present is mostly used
Distribution of all temporal-aspectual devices
%
60
40
20
0
4–5 yrs 7 yrs 10 yrs Adults
80
English German French
Chinese
Figure 10.4 Proportions of regional connectives across ages
at the adult age. In German the present mostly occurs with regional devices at all ages. In addition, regional devices also co-occur with thePr¨ateritumat seven years and with thePerfektat ten years. In both French and German the present mostly occurs with sequential devices. Chinese clauses that contain no aspect particles (NPCL) frequently contain regional devices, although there is an increase in the conjoined uses of these devices and of the particle le. Clauses without particles and with the particleleboth occur frequently with sequential devices at all ages, although conjoined uses involvingleincrease at the adult age.
Figure 10.5 further shows the overall proportions of inflections or particles that were used in conjunction with regional devices, while Figure 10.6 shows these proportions across ages for the most frequent forms within each language. In the Indo-European languages aspectually unmarked present inflections (all present in- flections in French and German, simple non-progressive ones in English) are less frequently used with regional devices than with sequential ones at all ages except at adult age. There is an increase in conjoined uses of regional devices and of past per- fective inflections (English simple past, French pass´e compos´eandpass´e simple, GermanPerfekt), although this increase occurs earliest in French (seven years on).
Time
Table 10.3 Connectives as a function of inflections and particles
4 –5 years 7 years 10 years Adults
Reg Seq Reg Seq Reg Seq Reg Seq
ENGLISH
PR .24 .26 .14 .33 .19 .42 .38 .63
PRPG .38 .06 .23 .03 .02 — .31 .11
PAST .24 .52 .51 .57 .74 .56 .15 .15
PAPG .05 .06 .08 .06 .05 — .07 —
OTHER .09 .09 .03 — — .02 .09 .11
FRENCH
PR 1.0 .80 .50 .91 .42 .79 .75 .64
PC/PS — .15 .25 .06 .25 .15 .08 —
IMP — .03 .25 .04 .25 .06 .17 —
OTHER — .01 — — .08 — — .36
GERMAN
PR .80 .53 .61 .82 .67 .87 .90 .81
PFK .10 .21 .08 .06 .22 .09 .10 .06
PTT — .13 .23 .11 .05 .02 — .06
OTHER .10 .13 .08 — .05 .02 — .06
CHINESE
LE .29 .27 .31 .34 .16 .32 .36 .45
IMP — .06 — .13 .04 .06 .08 .04
NPCL .71 .57 .69 .53 .80 .61 .56 .50
Among the rare imperfective inflections that occur, the English present progressive co-occurs with regional devices, rather than with sequential ones, from four/five years on. The conjoined use of regional devices with the Frenchimparfaitincreases across all ages, while conjoined uses increase until ten years (to decrease thereafter) in the case of the GermanPr¨ateritum. As for Chinese, regional devices increase with clauses containinglefrom four/five years on, but they increase even more with clauses containing no particle or imperfective markings from four/five and seven years on, respectively.