Aphasia research in Bahasa Indonesia

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
Validating A Test on Temporal and Aspectual Adverbs of Time
Denoting Past Tense to Non-Brain Damaged Speakers of Bahasa
Indonesia

1. INTRODUCTION
This paper reports development of a test on temporal and aspectual adverbs of time denoting past
tense in Bahasa Indonesia. The test is developed to be used with individuals with Broca's aphasia,
who have been shown by crosslinguistic research to be impaired in the production of tense. Of
interest is the fact that in the literature this deficit has been observed only in the languages that
inflect verbs for tense and that in Bahasa Indonesia no such inflection is found. Tense is marked by
either temporal or aspectual adverbs. The developed test reported in this paper can shed light to
what will happen in the agrammatic production of tense in Bahasa Indonesia compared to what has
been reported in the literature with speakers of languages that use inflections to mark tense.
According to Crystal (2003), tense is defined as a category used in the grammatical description of
verbs (along with aspect and mood), referring primarily to the way the grammar marks the time at
which the action denoted by the verb took place. Different languages have different number of
tenses and different ways of denoting the tenses. Most notably in Indo-European languages, tense is
shown by inflection attached to the finite verbs in clauses. This inflection has proven to be of
interest to linguists since crosslinguistic evidence of breakdown in producing and understanding the
tense inflection has helped to understand the processes involved in language production and
comprehension.

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
Friedmann and Grodzinsky (1997) reported a Hebrew-speaking patient who could produce
agreement inflection perfectly but was impaired in the production of tense inflection, use of copula,
and embedded structures. This deficit seems to be selectively observed in production as her
grammaticality judgment was significantly better.

Figure 1. A comparison of Friedmann and Grodzinsky's patient's performance on Judgment versus Production (source:
Friedmann and Grodzinsky 1997 p. 408)
To account for this finding, they proposed the Tree-Pruning-Hypothesis (TPH):
a. C, T, or Agr is underspecified in agrammatism.
b. An underspecified node cannot project any higher.
The possible deficits according to the TPH are Complementizer, Tense, and Agreement are
impaired, Complementizer and Tense are impaired, and only Complementizer is impaired.

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia

Figure 2. Agrammatic phrase marker. Arch represents site of deficit. (source: Lee 2003 p. 172)
However, some researchers have found patterns of deficits that are not in line with the TPH. Lee
(2003) in her study on Korean agrammatism, found out that the higher node (Mood in matrix
sentences and Complementizer in embedded clauses) is preserved while the lower node (Tense in
matrix sentences and Mood in embedded clauses) is impaired.

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia

Figure 3. The syntactic tree for a Korean clause (source: Lee 2003 p. 176)
Along the same lines, Wenzlaff and Clahsen's subjects (2004) were more impaired in the
production* and comprehension of tense inflection compared to the production and comprehension
of agreement inflection. Of special importance is the hierarchy of TP and AgrP mentioned by the
authors. They stated that in German AgrP is above TP and therefore impaired AgrP predicts
impaired TP, which is not supported by their finding. Based on this result, they proposed the Tense-
Underspecification-Hypothesis (TUH). The main point of the TUH is that AgrP and TP do not have
hierarchical positions and both of them are hosted by the T-node. In agrammatism, the T-node is
underspecified for tense features and as a result it predicts that Tense will be more impaired than
Agreement regardless of their hierarchy in the syntactic tree.
* This is mentioned following the term used by the authors although some people would possibly argue that the
production task carried out was not actually asking the subjects to produce the correct tense and agreement inflections
thus the results are not quite valid to state something about production.

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
The present author would like to draw attention to the speculation made by Burchert, Swoboda-
Moll, and De Bleser (2005) as to why Tense but not Agreement is underspecified in agrammatism
as this will serve as the basis of the discussion about tense in Bahasa Indonesia in the next section of
this paper. They speculated that "Tense requires the establishment of an anaphoric relationship
between the speech act and an event time in the discourse, unlike agreement marking. Such
discourse related phenomena have been found to be impaired in agrammatic aphasia, as was shown
in the interpretation of non-reflexive pronoun" (p. 192). One of the studies on non-reflexive
pronouns cited is that by Grodzinsky, Wexler, Chien, Marakovitz and Solomon (1993). In this
study, the agrammatic speakers had problems in applying Rule-i (NP A cannot corefer with NP B if
replacing A with C, C A-bound by B yields an indistinguishable interpretation). This rule is applied
in the following examples (Grodzinsky et al. 1993).
a. Is Mama Bear touching herself?
b. Is Mama Bear touching her?
Her in (b) cannot corefer with Mama Bear because this will result in a reading that is
indistinguishable from sentence (a). Grodzinsky et al. (1993) argued that the aphasics' failure to
apply Rule-i for sentences like (b) (i.e. to decide whether her corefers or not with Mama Bear) is
caused by their inability to hold the two representations in their working memory and to compare it
with context to come up with the best reading. This anaphoric relationship between context and
non-reflexive pronouns is the one assumed to be also responsible for the underspecification of
Tense in the TUH.
Before moving on to the report of the preliminary attempt, I would like to state the reasons why two
tasks are used in the preliminary attempt reported in this paper: repetition and reading tasks (see
Method section for more details). The reason why the two tasks are employed is to show whether
Burchert, Swoboda-Moll, and De Bleser’s (2005) speculation mentioned earlier is due to Tense per
se or a result of the task used (discourse/production). Furthermore, reading task makes sure that the

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
participants really process the language and not just parroting back the experimental sentences. Of
course, this design will affect the choice of aphasic participants in the later part of the research. The
aphasic patients need to have relatively minor problems regarding reading and repeating.
Below I sketch the structure of Bahasa Indonesia that is tested and state the research questions in
light of the previous discussion based on the literature.
2. THE REALIZATIONS OF PAST TENSE IN BAHASA INDONESIA
Bahasa Indonesia is the national language of Indonesia, spoken by approximately 200 million
people. Only about one-third of the number (700 thousand) acquire it as a mother tongue as the
majority of the Indonesian population still speak one or more of the languages of their respective
ethnic group. Even among those who acquire it as a mother tongue, dialectal differences can be
observed that mainly show the influence of the ethnic languages of parents or other adults
interacting closely with the children A possible condition is that children of Javanese parents born
and raised in Jakarta, the capital, more often than not acquire Bahasa Indonesia as their mother
tongue although the vocabulary and structure often reflect the native language(s) of their parents or
other care givers and in turn their language will be different from that of children whose parents are
Sundanese or from the eastern or more western parts of the country.
Bahasa Indonesia is an SVO language. The verbs in Bahasa Indonesia take the same form for all
numbers and persons. This also applies for tense. Tense is not denoted by inflection attached to the
verbs but by temporal and aspectual adverbs of time. Aspectual adverbs of time (sudah and telah)
always come before the main verbs while the temporal adverbs can come at the end of sentences
(default position), before the subject, and after the subject. The following sentences are examples of
past tense sentences taken from the developed test.

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
a. Dia pergi ke rumah sakit kemarin .
(S/he went to the hospital yesterday.)
b. Kemarin Ani menyelesaikan pekerjaan rumahnya.
(Yesterday Ani finished her homework.)
c. Mereka kemarin lusa membuat sebuah rencana besar.
(They two days ago made a big plan.)
d. Mereka sudah memasak makanan untuk sarapan.
(They have cooked the food for breakfast.)
e. Saya telah meminta izin dari kantor.
(I have asked permission from the office.)
However, some verbs in Bahasa Indonesia are inflected to show transitivity. In the folowing
examples, sentence (f) is an example of an intransitive sentence using the verb jatuh (fall) while
sentence (g) is its transitive counterpart.
f. Gelas itu jatuh.
(The glass falls/is falling.)
g. Tiwi menjatuhkan gelas itu.
(Tiwi drops the glass.)

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
Going back to past tense in Bahasa Indonesia, the following trees are proposed for sentences (a) to
(e). S

NP VP

Pron VP NP AdvP

Saya Adv V N Adv N
telah meminta izin dari kantor
Figure 4. The syntactic tree for a past tense sentence with aspectual adverb telah
 S
NP VP

 Pron VP

Dia

 AdvP AdvP
V AdvP NP Adv
pergi Adv N Adj kemarin
ke rumah sakit
Figure 5. The syntactic tree for a past tense sentence with temporal adverb kemarin in its default position
As can be observed from figures 4 and 5, verbs do not need to collect or to be checked for tense in
Bahasa Indonesia. In light of the previous discussion about agrammatic production in the languages
that inflect verbs for tense, it is interesting to ask what will happen in Bahasa Indonesia. The first
possibility is that there can be a dissociation in the production of the two kinds of adverbs. The
aspectual adverbs telah and sudah are not anchored at a certain time. What they show is that the
actions happened at an unspecified time in the past. On the other hand, the temporal adverbs (e.g.
kemarin, kemarin lusa) explicitly give the time of happening. Also, syntactically sudah and telah

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
have their fixed position in sentences (i.e. before the finite verbs) while the temporal adverbs can
occupy one of three possible positions (see example sentences (a) to (c)). The interplay of the
physical and syntactic properties of the two kinds of adverbs may lead to dissociation in their
production by Broca's aphasics. One kind of adverb may be produced quite successfully while the
production of the other one may be significantly impaired. The temporal adverbs may be easier for
the aphasic patients to produce because they are anchored at specific points in time and hence do not
engage memory very much. However, their three possible positions in sentences may pose
problems. On the other hand, the aspectual adverbs do not specify points in time but their positions
are more fixed in sentences. This combination may also lead to the aphasic patients’ success in
producing the aspectual adverbs as compared to that of the temporal adverbs.
The second possibility is the total failure to produce the adverbs. The reasoning follows Burchert,
Swoboda-Moll, and De Bleser's (2005) speculation on why Tense but not Agreement is
underspecified in agrammatism. This underspecification of Tense-features is because there is a
difference in the semantic interpretability between Tense and Agreement. Tense is underspecified
because it is semantically interpretable and taxes the working memory of the aphasic patients
(Burchert, Swoboda-Moll, De Bleser (2005). If Bahasa Indonesia-speaking Broca's aphasics have
problems in holding in memory the past tense sentences read to them and to differentiate the now
and here of the test administration and the there and then mentioned in the sentences, they may
abandon the adverbs altogether and produce present tense sentences in an attempt to repeat the
sentences (for details of the test, see the Method section). One piece of evidence that can support
this possibility is Friedmann’s (2000) finding that her Hebrew-speaking participants substituted one
tense with another in a repetition task. Nevertheless, something different might happen in the
reading task as memory is not so much demanded here compared to the condition in the repetition
task. Perhaps dissociation or failure is observed to a lesser degree in the reading task.

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
In summary, the validated test reported in this paper is developed to assess the production of
temporal and aspectual adverbs of time in the speech of Bahasa Indonesia-speaking agrammatics. It
aims at answering three questions:
1. Is the success in reading and repeating adverbs marking past tense in Bahasa Indonesia by
aphasic patients influenced by different working memory engagement poses by the repetition and
reading tasks?
2. If yes, which kind of adverb is more difficult to produce in each of the task?
3. What factors may cause this observed pattern of difficulty?
3. METHOD
3.1. Subjects
Ten non-brain damaged Bahasa Indonesia native speakers without hearing defect and history of
dyslexia volunteered in the validation of the test. Seven of them are males. As true native speakers
of standard Bahasa Indonesia are very hard to be found based on the explanation in section 2, the
experimenter/author tried to make sure that the volunteers were exposed to a form of Bahasa
Indonesia when they were young children if not from birth by asking them beforehand. They are
master students studying in The Netherlands (9 in Groningen and 1 in The Hague) with age range
between 25 to 40 (mean age 33).

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
3.2. Materials
The subjects were presented 30 sentences of standard Bahasa Indonesia in the past tense. The
sentences were presented orally one by one in the repetition task and were shown one by one on a
piece of hard paper in the reading task. Fifteen of those are sentences with aspectual adverbs and the
rest are sentences with temporal adverbs (see table 1 for some examples and Appendix 2 for a
complete list). Seven of the sentences with aspectual adverbs have sudah as the adverb marking past
tense and the rest have telah. The sentences with temporal adverbs are divided further based on the
position of the adverbs in the sentences: five with the adverbs at the beginning of the sentences, five
with the adverbs located after the subjects, and the other five with the adverbs at the end of the
sentences (default position). The sentences were randomized so that a sentence with a temporal
adverb does not come after another sentence that also has a temporal adverb.
There are two tasks: a repetition task and a reading task. The same set of sentences is used in the
two tasks to ensure comparison in the subjects's performance. After finishing the two tasks, the
participants were asked to volunteer their opinion on the grammaticality of the sentences.

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia

Sentences with aspectual adverbs Sentences with temporal adverbs
Saya sudah bertemu dia di rumah.
(I have met him/her at home.)
Tadi pagi Ari menyetujui usul saya.
(This morning Ari agreed to my suggestion.)
Dewi telah mengirim surat kepada ibunya.
(Dewi has sent the letter to her mother.)
Saya tadi pagi meminta izin dari kantor.
(I this morning asked permission from the
office.)
Adam telah membaca sebuah novel terkenal.
(Adam has read a famous novel.)
Saya bertemu dia di rumah tadi pagi.
(I met him/her at home this morning.)
Table 1. Examples of experimental sentences used in the repetition and reading tasks
3.3 Procedure
In the repetition task, after practice with the example sentences, the experimenter read out the
experimental sentences one by one and each of the subjects was asked to repeat the sentences. On
request, the experimenter read out the sentences for a second time and the subjects were given one
chance to repeat each of the sentences. In the reading task administered a few hours after the
repetition task, the same sentences were used. To minimize distraction, each of the sentences is
written on a piece of hard paper and, after practice with example sentences, the subjects were asked
to read the sentences aloud one by one.
3.4. Scoring
A simple correct/incorrect system was used based on whether the subjects produced the adverbs in
the correct position as mentioned in the experimental sentences. The errors were then analysed

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
qualitatively. The correct number of sentences with aspectual adverbs produced was compared with
the correct number of sentences with temporal adverbs produced. Then for the sentences with the
temporal adverbs, errors were checked whether they occured more in the sentences with the adverbs
at the beginning of the sentences, in the sentences in which the adverbs come after the subjects, or
in the sentences with the adverbs at the end of the sentences. Furthermore, errors were classified as
omission, inflection, semantic paraphrasia, or other.
4. RESULTS
The subjects performed perfectly (100%) in the reading task while in the repetition task, the mean
percentage of correctly repeated sentences is 96.3%. However, this ceiling performance may have
been due to their literacy level and education as the participants are all graduate students. Overall,
there were 11 mistakes, the majority of which (8) involved adding the aspectual adverb sudah or
telah to a sentence that already contains a temporal adverb resulting in sentences like "*Mereka
kemarin lusa telah membuat sebuah rencana besar." The other three mistakes involved substitution
(substituting telah with sudah), addition (adding sudah to a sentence that already contains telah) and
word order (Kami kemarin instead of Kemarin kami). The addition of telah or sudah to sentences
that already contain temporal adverbs seemed to be a result of interference of the previous
sentences. The eight mistakes involved five different sentences and each of those sentences are
preceeded by a sentence with telah or sudah. Furthermore, six of the eight additions of telah or
sudah happened to sentences that have the pattern Subject + Temporal Adverb + Verb. The other
two happened to sentences with the patterns Temporal Adverb + Subject + Verb and Subject +
Verb + Temporal Adverb respectively.

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
However, the explanation that the ungrammatical addition of telah or sudah happened only because
of interference is not adequate since this phenomenon does not affect the three patterns to the same
degree. The sentences whose pattern is Subject + Temporal Adverb + Verb seem to be more
vulnerable to the ungrammatical addition of sudah or telah since 3 out of 5 sentences with this
pattern preceeded by telah or sudah are affected, compared to 1 out of 2 with the pattern Temporal
Adverb + Subject + Verb and 1 out of 3 with the pattern Subject + Verb + Temporal Adverb.
This vulnerability of the pattern Subject + Temporal Adverb + Verb is also coupled by the fact
that 30% (3 out of 10) of the participants judged this pattern as the strangest or least common and
hence the most difficult to produce. Each of the other two patterns (Temporal Adverb + Subject +
Verb and Subject + Verb + Temporal Adverb) is judged as the strangest or least common only
by 10% of the subjects. This judgment and the error patterns mentioned previously seem to explain
the vulnerability of this pattern to the ungrammatical addition of telah or sudah.
Pattern No. of sentences
preceeded by a sentence
with telah or sudah
No. of sentences
ungrammatically added
with telah or sudah
Percentage of subjects
judging the pattern as the
strangest
Subject + Temporal
Adverb + Verb 5 3 30%
Temporal Adverb +
Subject + Verb 2 1 10%
Subject + Verb +
Temporal Adverb 3 1 10%
Table 2. Summary of errors in the repetition task and subjects' judgment on patterns of sentences with temporal adverbs


Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
The aforementioned errors did not happen in the reading task as all of the subjects performed
perfectly. This is perhaps due to the fact that the subjects did not have to memorize the words as
they did in the repetition task. Some of the subjects said that in terms of difficulty, the reading task
was easier because they could have their own pace and did not have to memorize the sentences.
However, this condition did not alter their judgment on the sentences as they still rated the patterns,
especially those with the temporal adverbs, the same way as they did after they completed the
repetition task.
In order to use this test with agrammatic speakers, some sentences need to be changed as suggested
by the subjects. Eight sentences containing the word itu (a demonstrative pronoun that acts like the
in English) are changed due to a lack of context that does not support the definiteness of the nouns.

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia

Original sentences Revised sentences
Kami telah menjual mobil sedan itu.
(We have sold the sedan.)
Kami telah menjual sebuah mobil sedan.
(We have sold a sedan.)
Kami membicarakan masalah itu minggu lalu.
(We talked about the problem last week.)
Kami membicarakan masalah ibu minggu lalu.
(We talked about mother's problem last week.)
Kami kemarin menjual mobil sedan itu.
(We yesterday sold the sedan.)
Kami kemarin menjual sebuah mobil sedan.
(We yesterday sold a sedan.)
Ari sudah menyetujui usul saya itu.
(Ari has agreed to my suggestion.)
Ari sudah menyetujui usul saya.
(Ari has agreed to my suggestion.)
Kami sudah membicarakan masalah itu.
(We have talked about the problem.)
Kami sudah membicarakan masalah ibu.
(We have talked about mother's problem.)
Tadi pagi Ari menyetujui usul saya itu.
(This morning Ari agreed to my suggestion.)
Tadi pagi Ari menyetujui usul saya.
(This morning Ari agreed to my suggestion.)
Tim itu telah memenangkan partai final.
(The team has won the final match.)
Persija Jakarta telah memenangkan partai final.
(Persija Jakarta has won the final match.)
Tim itu bulan lalu memenangkan partai final.
(The team last month won the final match.)
Persija Jakarta bulan lalu memenangkan partai
final.
(Persija Jakarta last month won the final match.)
Table 3. Revised sentences based on the feedback from the subjects
In addition, a sentence will be changed due to something some of the subjects called "illogical." The
sentence is "Mereka memasak makanan untuk sarapan bulan lalu (They cooked the food for
breakfast last month)." The original intention was the temporal adverb bulan lalu (last month) is to
modify the verb memasak (cook). However, putting bulan lalu right after sarapan (breakfast) made

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
it sound that bulan lalu modifies sarapan which they said strange because breakfast is served in the
morning. However, changing bulan lalu with another temporal adverb still risks the chance of
reading the temporal adverb is an adjective that modifies sarapan although it is actually meant to
modify the verb memasak. Therefore, it was decided to change the part of the sentence before the
temporal adverb and leave the temporal adverb as it is. The new sentence is "Rini melahirkan anak
pertamanya bulan lalu (Rini gave birth to her first child last month)." Another change also involves
a kind of ambiguity. The sentence in question is "Ayah membeli rumah baru kemarin lusa" (Father
bought a new house two days ago.) One of the subjects said that the reading can depend on how we
cluster baru (new). One possibility is baru clusters with rumah thus forming a noun phrase (new
house). The other possibility is baru clusters with kemarin lusa thus forming an adverb phrase (just
two days ago). As a result, it was decided to change the adjective baru with besar (big) that is only
possible to be clustered with rumah thus forming a noun phrase (big house) as was originally meant.
Another issue that surfaced in the discussion with the subjects was the issue of register. Since the
sentences in the test are all in standard Bahasa Indonesia, some of them said that they are not very
suitable for the repetition task. The reason was because standard Bahasa Indonesia is usually found
in written and not spoken language. In response to this comment, I would like to state that using
standard Bahasa Indonesia in the test is a kind of compromise because there is no way I could know
in advance what dialect of Bahasa Indonesia my subjects would speak. If I were to take into account
the dialect(s) each of my subjects speaks, then I would have to construct a different test for each of
them. Apart from the fact that I do not know very much about the structural characteristics of all the
dialects spoken in the country, these different tests will be difficult and time consuming to validate.
This, I believe, is work for the future. For now, I deal with the problem by changing the instruction
read before the two tasks of the test. The instruction becomes:

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
1. In the repetition task " Saya akan membacakan sebuah kalimat dalam Bahasa Indonesia resmi.
Mohon kalimat tersebut diulang." (I will read you a sentence in standard Bahasa Indonesia. Please
repeat the sentence.)
2. In the reading task "Saya akan menunjukkan sebuah kalimat dalam Bahasa Indonesia resmi.
Mohon kalimat tersebut dibaca keras." (I will show you a sentence in standard Bahasa Indonesia.
Please read the sentence aloud.)
In summary, three findings resulted from the validation activity and some changes have been made
to accommodate the subjects' comments. The first finding is that the aspectual adverbs sudah and
telah are not difficult to read and repeat as no deletion to these adverbs was observed. This is in line
with the finding of Druks and Carroll (2005) that the more successful –ing form production of their
subject as compared to his production of tensed verbs was because –ing shows aspect and not tense.
The second finding is among the three positions of temporal adverbs, the position after the subject is
the most vulnerable to ungrammatical addition of sudah or telah. Remembering that the temporal
adverbs denote tense “purely” compared to the aspectual adverbs, this finding is in line with what
has been observed with languages that inflect verbs for tense. The third finding is that the reading
task was easier for the participants than the repetition task as shown by their higher percentage of
correct response in the former task. In total, eight sentences were changed to accommodate the
subjects' comments. The instruction was also revised to ensure the match between it and the test
material. The revised instruction and sentences can be found in Appendix 1 and the sentences with
their English translation in Appendix 2.
The result of the preliminary attempt to develop the test on temporal and aspectual adverbs of time
denoting past tense in Bahasa Indonesia reported in this paper seems to show that tense in this
language is also vulnerable in the reading and repetition of aphasic patients. The observed pattern
that more mistakes were produced in sentences with the temporal adverbs indicates that aspect

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia
denoting past time is less vulnerable. Therefore, it appears that it is not tense per se that is
vulnerable in Bahasa Indonesia but it depends on how it is realized in sentences. As agreement is
not marked in Bahasa Indonesia, this result does not support either of the hypotheses mentioned
previously (the Tree-Pruning-Hypothesis or the Tense Underspecification Hypothesis). In addition,
this result can predict the aphasic performance. If the non-brain damaged speakers with properly
working working-memory and language system give evidence that the temporal adverbs are more
difficult in the two tasks, then the aphasic patients with limitations in working memory and
language system can be expected to have more problems with the temporal compared to the
aspectual adverbs.
5. CONCLUSION
It has been shown by several studies that agrammatic patients are impaired in the production of
inflectional morphemes denoting tense. Some hypotheses have been put forward to account for the
findings. The test reported in this paper is developed to find out what will happen in Bahasa
Indonesia regarding past tense. In this language past tense is marked by temporal and aspectual
adverbs of time, unlike the languages that show tense by inflections attached to verbs so far reported
in the literature. Some improvements have been made to the test material as a result of the opinions
volunteered by the subjects. The validated test is thus ready to be used with agrammatic patients in
order to answer the three research questions posed.

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia

REFERENCE

Avrutin, S., & Wexler, K. (2000) Children's Knowledge of Subjuntive Clauses: Obviation, Binding,
and Reference. Language Acquisition, 8 (1), 69-102.
Burchert, F., Swoboda-Moll, M., & De Bleser, R. (2005). Tense and Agreement dissociations in
German agrammatic speakers: Underspecification vs. hierarchy. Brain and Language, 94, 188
199.
Crystal, D. (2003). A Dictionary of Linguistics & Phonetics (fifth edition). Malden: Blackwell
Publishing.
Druks, J. & Carroll, E. (2005) The crucial role of tense for verb production. Brain and Language,
94, 1-18.
Friedmann, N., & Grodzinsky, Y. (1997). Tense and Agreement in agrammatic production: Pruning
the syntactic tree. Brain and Language, 56, 397-425.
Friedmann, N. (2000). Moving verbs in agrammatic production. In Bastiaanse, R., & Grodzinsky,
Y. (eds.) Grammatical Disorders in Aphasia: A Neurolinguistic Perspective. London: Whurr.
Friedmann, N. (2002). Question production in agrammatism: The Tree Pruning Hypothesis. Brain
and Language, 80, 160-187.
Grodzinsky, Y., Wexler, K., Chien, Y.-C., Marakovitz, S., & Solomon, J. (1993). The breakdown of
binding relations. Brain and Language, 45, 396-422.
Lee, M. (2003). Dissociations among functional categories in Korean agrammatism. Brain and
Language, 84, 170-188.
Ruigendijk, E., & Avrutin, S. (2003). The comprehension of pronouns and reflexives in agrammatic
and Wernicke's aphasia. Brain and Language, 87, 17-18.
Saeed, J. I. (2003). Semantics (second edition). Malden: Blackwell Publishing.
Sakri, A. (1995). Bangun Kalimat Bahasa Indonesia (edisi kedua). Bandung: Penerbit ITB.
Wenzlaff, M., & Clahsen, H. (2004). Tense and Agreement in German agrammatism. Brain and
Language, 89, 57-68.

Harwintha Y. Anjarningsih 1572512 EMCL 2005-2006 Crosslinguistic Study to Aphasia