Schema theory provides a meeting place for the studies of language, culture, and cognition. Cultural knowledge and experience are represented in the form of cognitive schemas, which underlie the production of cultural discourse. Thus, an examination of distinctive patterns of discourse produced by cultural groups may provide us with an understanding of their underlying cultural schemas. Discourse in English produced by Aboriginal students reveals certain Aboriginal cultural schemas. A closer analysis of Aboriginal English discourse also reveals some salient patterns which may be explained by the way these cultural schemas are processed. The present study is an initial attempt to explore these patterns in the light of schema theory. In this analysis, certain features suggest distinctive patterns of schema activation, while some other point to the possibility of a distinctive role played by schemas in the use of referential devices.

Schema theory
Cognitive blocks used for the organisation of information are called schemas (Rumelhart, 1980). Nishida (1999) defines schemas as "generalized collections of knowledge of past experiences which are organized into related knowledge groups and are used to guide our behaviours in familiar situations" (p. 755).

Schema theory maintains that information processing is mediated by mental structures that organize related pieces of knowledge. Schema theory has proved to be of high explanatory power in cognitive studies for more than half a century. (Bobrow & Norman, 1975; Minsky, 1975; Rumelhart, 1980; Schank & Abelson, 1977). "It [schema theory] has been used by both cognitive and social psychologists to explain a wide array of phenomena concerning accuracy and distortion in both perception and memory" (Koriat, Goldsmith & Pansky, 2000, p. 494).

The theory has, however, undergone a change in the way it is modeled alongside the paradigm shift in cognitive science (i.e., classicism to connectionism). In connectionist models, a schema is not a set of sentences, but rather a pattern of interaction among strongly interconnected units (Rumelhart, Smolensky, McClelland, & Hinton, 1986).
 

Current connectionist models of cognition give us a much less rigid way of understanding schemas and the meanings to which they give rise ... Meanings generated by schemas, in connectionist models, are mental states but are shaped by the learner's specific life experiences and are sensitive to activity in a particular context. While often similar from person to person, context to context, and one period of time to another, they can vary and change. (Strauss & Quinn, 1997, p. 50)

Schema theory has been employed by researcher across several disciplines such as artificial intelligence (Arbib, 1994), cognitive linguistics (Langacker 1987, 1991; Chafe, 1994), cultural linguistics (Palmer, 1996), and cognitive anthropology (Strauss & Quinn, 1997: D'Andrade, 1995). Within the framework of cognitive linguistics, Langacker (1987) uses the term schema to refer to a kind of pattern (e.g., CVC) or template which a speaker acquires through exposure to multiple exemplars of the pattern.
In developing a theory of cultural linguistics, Palmer (1996) employs the concept of schema and maintains that "[i]t is likely that all native knowledge of language and culture belongs to cultural schemas and the living of culture and the speaking of language consist of schemas in action" (p. 63).

Cognitive anthropologists view culture as a cognitive system and therefore draw on cognitive theories, such as schema theory, to enhance their understanding of the way culture may be represented in the mind. Stauss and Quin (1997), for example, have drawn on connectionism and schema theory in their studies of cultural meanings. Strauss and Quinn believe that schemas could be predetermined genetically or else they could be cultural. They maintain that "culture ... consists of regular occurrences in the humanly created world, in the schemas people share as a result of these, and in the interactions between these schemas and this world" (p.7).

The merit of schema theory lies in the interface that it provides between the studies of culture, cognition, and language. This is due to the fact that schemas are cognitive structures that can be determined by cultural experiences and are reflected in linguistic expression. In other words, schema theory provide a powerful tool for examining culturally determined cognitive structures which underlie the production of distinctive discourse patterns.

Schema theory can shed some light on the factors which may either facilitate or debilitate cross-cultural communication. As Malcolm, et al. (1999) put it, "[t]he pervasiveness of schemas in approaching and interpreting experience makes them a key element in communicating inclusion or exclusion in cross-cultural communication" (p. 74). Malcolm (Malcolm, et al, 1999; Malcolm & Rochecouste, 2000) has recently applied schema theory to the study of semantic aspects of Aboriginal English. Before presenting Malcolm's analysis, a brief description of the varieties of English known as Aboriginal English seems to be in order.

Aboriginal English
Aboriginal English is a cover term referring to dialects of English, except Standard Australian English, spoken by Aboriginal Australians. For many speakers of Aboriginal English, this dialect is a successor of Aboriginal languages that are largely extinct now (Malcolm, 2000). This is due to the fact that Aboriginal English includes features from these Aboriginal languages, as well as from English. Of course, Aboriginal English includes features that belong neither to the former nor to the latter. Malcolm (1995) defines Aboriginal English as:
 

A range of varieties of English spoken by many Aboriginal people and some others in close contact with them which differ in systematic ways from standard Australian English at all levels of linguistic structure (sounds; word forms; syntax; vocabulary; meanings) and which are used for distinctive speech events, acts and genres. (p. 19)

Although Aboriginal English is not a pidgin or creole now, there is evidence that it has gone through the processes of pidginization, creolization and decreolization and is mainly used now for intra-communication purposes (Malcolm, 2000; also see Malcolm & Koscielecki, 1997, for a diachronic study of Aboriginal English). Aboriginal English is different from Australian English both at the surface level of formal features and also the deeper level of semantic content (Arthur, 1996; Harkins, 1994; Malcolm et al, 1999). Aboriginal English employs distinctive means to represent reality (Malcolm, 1994a, 1994b). Among these are:

a) Aboriginal English achieves economy of expression
b) Aboriginal English is highly context dependent
c) Aboriginal English foregrounds aspect, duration, dual number, participant relations, and oral art and backgrounds
    gender, existence, and plurality
d) Aboriginal English discourse reveals unique rhetorical structures.

Aboriginal English achieves certain unique functions for its speakers, such as a) creating a convivial atmosphere among Aboriginal speakers, b) reinforcing common Aboriginal Identity c) providing for certain Aboriginal genres and d) achieving ironic humor (Malcolm, 1995). Research has shown that "even where Aboriginal English seems to employ the same vocabulary as Australian English, it is informed by a semantics deeply rooted in Aboriginal culture" (Malcolm & Rochecouste, 2000, P. 98). This has often resulted in miscommunication between Aboriginal children and non-Aboriginal teachers (Malcolm, 1977, 1982). As Malcolm, et al. (1999) put it:
 

We have seen that the same English words and expressions can accommodate contrasting cultural schemas, so that speakers of standard English may think (on the basis of surface linguistic form) they are being understood by Aboriginal English speakers (and vice versa) but may be drawing on completely different inferences from the communication from those which were intended. (p. 74)

The above observation served as the incentive for the application of schema theory to the task of identifying cultural schemas which inform Aboriginal English discourse.

Schemas reflected in Aboriginal English discourse
By comparing 40 passages and determining their genre-specific recurring content features and discourse patterns, Malcolm has been able to identify a number of distinctive Aboriginal cultural schemas (Malcolm, 2000; Malcolm & Rochecouste, 2000). The labels he has used for these schemas are: Travel, Hunting, Observing, Encountering the Unknown, Gathering, Isolation from the Group, Problem Solving, and Borrowed schemas. Malcolm observes that the first four are the most frequently occurring in the data that he has analyzed. He defines these schemata as follows:

a) Travel Schema:
Experience of known participants organized in terms of alternating
travelling (or moving) and non-travelling (or stopping) segments,
usually referenced to a known time of departure and optionally
including a return to the starting point.

b) Hunting Schema:
Experience of known participants organized with respect to the
observation, pursuit and capture of prey, usually entailing killing and
sometimes eating it. Success is usually associated with persistence.

c) Observing Schema:
Experience - usually shared experience - recalled in terms of details
observed in it, whether of natural or social phenomena. The recall
may function to provide a check on - or give evidence of - learning.

d) Encountering the Unknown Schema:
Experience (either first-hand or vicarious) of strange powers or
persons affecting normal life within the community.
(Malcolm & Rochecouste, 2000, pp. 103-104)

Certain discourse markers and discourse strategies appear to be associated with these schemas in Aboriginal English discourse (see Malcom, 2000). The analysis of these texts also reveals certain recurrent features which point to some distinctive patterns in the processing of the above-mentioned cultural schemas. This analysis is based on examining different aspects of more than 100 texts, both written and spoken, produced by Aboriginal students from different age groups living in Western Australia and the Northern Territory. The following sections highlight aspects of Aboriginal schematic processing which stand out in recurring patterns in the data analyzed for this study.

Aboriginal schematic processing
The analysis of the texts in Aboriginal English suggests that during the processing of discourse, sometimes schemas trigger one another in an exponential fashion. That is, consecutive sentences in a text may be attributed to different Aboriginal schemas. In fact several schemas seem to underlie the production of a relatively short piece of discourse. Frequently, the activation of a new schema is not marked by any discourse markers. The following story written by an Aboriginal student may exemplify this pattern.

Text 1
1 like doing washing at Camp
2 and my name is Rosalyn
3 I live at McLaren
4 and my brother Peter Stevens Ricky Simon
5 and sisters Josephine Sussanne Lorraine Jane Lisa Mirelle
6 and my sisters and brothers tell Some Story for me
7 and they always went hunting for Emo and Truckey
8 I like to drive bus

As evident, this story is something of an autobiography, in which the author introduces herself (lines 2-3) and the members of her family (lines 4-5) and then mentioning the names of the family members activates Story Telling (line 6) and Hunting (line 7) schemas. The activation of Hunting schema in turn triggers Bus Driving (line 8), a Travel schema, which is a subschema of the author's Hunting schema.

This pattern of schema activation also suggests the existence of cultural associations such as family with hunting and hunting with bus driving. Bus driving can be reasonably associated with Hunting schema in Aboriginal children. This is because usually Aboriginal people have to travel a long way to get to the hunting ground and this could be by bus and that would usually be a good chance for Aboriginal children to learn how to drive. Patterns of association like this may make Aboriginal discourse less transparent to non-Aboriginal people. Schemas or sub-schemas which are associated with each other in Aboriginal cognition may not necessarily be considered as related by non-Aboriginal speakers and that is where culture and life experience leave their footprints on human cognition.

Another distinctive feature of discourse in Aboriginal English is the pattern observed that pieces of a text refer to different topics and one topic is introduced and reintroduced in the same text. It is as if a schema gets activated and then inactivated and the after a few sentences it is reactivated. It may also be that case that a previous schema has been kept activated but not verbally processed and then the person refocuses on that schema.

Text 2
1- Oh. .um my uncle David
2- he's at the station
3- and he was driving along
4- from Windmill run..
5- cos he drives that slow..
6- and he still comin back..
7- comin back ome..
8- an e's goin to this one windmill..
9- big mob o emus..
10- seen it..
11- y' know dey got pad
12- goin along de road..
13- an e was drivin along
14- an e seen
15- he looked in is revision mirror
16- and big light was
17- behin im..
18- so e didn't worry about it
19- e jus kept on goin along..
20- an when e looked again it wasn't dere..
21- it was in front of im
22- den.. e was driving along
23- and e got to one windmill
24- and bi-i-iggest mob o emus..
25- packed..
26- at that emu..
27- yeah you make emu farm outta dat..
28- yeah about sixty...seventy
29- Dey jus all scatter
 

In the above text, the narrator starts talking driving (Travel schema) and then seeing emus (Observing schema) and from line 14 to 21 the text is about some kind of light*, which is associated with Scary things schema. As of line 21, the narrator seems to refocuses on, or reactivate, the previous schemas of driving and observing. The repetition of "going along" (lines 10 &15) and "driving along" (lines 11 & 18) seem to maintain the Travel schema in the background.

Text 3
1- On the way back from Mullawa
2- We saw a goanna
3- And my um uncle he was hitting it on the head with a rock
4- and we caught it
5- and we ate it up for dinner
6- and um it tastes nice
7- and we went to Geraldton
8- and we had lots of fun
9- and we had to go 'ome for dinner
10- and my uncle ate most of it
11- He had it knock it on the head twice
12- And it was really fun....
13- Bye.

In the above text, the pattern discussed in the preceding section is evident in the repetition of "dinner' event (lines 9 and 5) and also "knocking the snack on the head" (lines 3 and 11). The narrator seems to have finished the story by line 8, but the next sentence (line 9) refocuses on "dinner" and line 11 also refocuses on "knocking the snack on the head" previously mentioned in line 3.

Schema-Based referencing
Another salient feature of the Aboriginal English texts analyzed for this study is may be called schema-based referencing. There are some pronouns or demonstratives in Aboriginal English discourse which reveal no anaphoric or cataphoric referencing functions. It seems that these referential devices retrieve their antecedents from activated schemas or images in the mind of the person, rather than the discourse preceding or following their reference (i.e., linguistic context). In other words, their referencing seems to be schema-oriented or image-oriented rather than discourse-oriented. These pronominal and demonstrative devices do not even seem to refer to any referents in the immediate context of the discourse (i.e., physical context) either. The following discourse excerpts would exemplify this feature of Aboriginal discourse processing:

Text 4
1- Oh um my uncle Patrick he's at the station
2- and he was driving along from Windmill run,
3- cos he drives that slow,
4- and he still coming back,
5- comin back ome,
6- and e's goin to this one windmill,
7- Big mob o emus,
8- Seen it,

Text 5
G: Michael e's just on the farm out there
EH: Where I' is here?
G: E- e's out the B- farm
EH: He came back
G: I think e went back down there for that funeral
 

Text 6
1 J My sister she went to.. she went to put er rubbish away..
2 K Yeah
3 J an she was messin roun
4 playin like dat
5 an.. she was gunna git cut
6 K mmm
7 J she came out like dat dere...
8 she nearly killed dat snake
9 A she stepped on the snake like dat
10 J Yeah.
 

It can be noted that the underlined demonstratives and pronouns in the above texts do not seem to refer to any elements in the discourse preceding or following their occurrence. It is as if the speaker is visualizing the scene or the schema and is making a reference, using demonstrative or pronominal devices, to the elements of these scenes and schemas, rather than tracing back the antecedents in the preceding discourse. Sometimes, such patterns of referencing seem to be pertinent to the assumption of shared cultural schemas. That is, referential devices may refer to default elements in Aboriginal cultural schemas. This is however not always the case, since in many cases the addressees in the conversations are either Aboriginal or non-Aboriginal research assistants who are not from the same regional locations as the students.
 

Cursory processing and activation
The texts analyzed for this study reveal another salient recurrent feature and that is the prevalence of minimal discourse. I use minimal discourse to refer to discourse marked by the pervasive use of elliptic utterances denoting and describing complete propositions and events. Verbal interaction in Aboriginal English is heavily marked by minimal discourse. Malcolm (1982) refers to this feature of Aboriginal English discourse as unelaborated response. The breakdown in communication between AE speakers and non-Aboriginal speakers may partly be attributed to this feature of Aboriginal English discourse. The following fragment of Aboriginal discourse exemplifies this phenomenon:

Text 7
1 No big boy A.. reckon
2 he help K..
3 was drivin back from Wiluna or whatever some place
4 an light behind,
5 look in r'vision mirror
6 no he's gone,
7 drivin along
8 saw i',
9 look in the 'vision mirror again,
10 look in the back seat,
11 an ole ole blackfella sittin in the back seat, lookin at im
 

Using the terminology of network models of memory (e.g., Sharifian & Samani, 1997), this discoursal feature may be attributed to the amount of activation reaching the nodes during the processing of discourse. It may also be the case that the nodes are activated in the memory network but not verbally processed (i.e., verbalized). The latter may stem from the assumptions made by the speaker about the audience's knowledge. That is, the assumption of shared cultural schemas made by the speaker can make complex verbal processing unnecessary. For example, in text 6, light behind (line 4) activates a schema that is shared by Aboriginal people coming from the same cultural background, and therefore makes it unnecessary for the speaker to elaborate on her sentences. This is an example of the case where certain texts engages schema-driven processing in the interlocutors who share their cultural schemas, while engaging data-driven processing in those who do not.

Conclusion
This paper highlighted some conspicuous features of discourse in Aboriginal English, suggesting some possible explanations, drawing on schema theory. Thus far, the application of schema theory to the study of the conceptual system underlying the use of Aboriginal English seems promising. Schema theory appears to be able to elucidate several aspects of discourse in Aboriginal English, drawing on the cultural knowledge which feed the conceptualizations underlying the formation of discourse by Aboriginal people. The pedagogical implications of this line of research is best articulated by Malcolm and Rochecouste (in press) as follows:
 

Aboriginal people are heirs to a rich, orally transmitted cultural tradition. The oral skills which are imparted to them as children both embody Aboriginal culture and provide them with the means of maintaining it. These skills are, for many Aboriginal children, the primary means of approaching experience and knowledge, that is, they are fundamental to learning. Despite this, they are not usually enabled to play any part in formal education. It is hoped that, by making Aboriginal children's discourse forms and their underlying schemas more explicit, it will be possible to enable more accommodating approaches to education to be implemented. (p. 2)

Arbib, M. (1994). Schema theory: Cooperative computation for brain theory and distributed AI. In V. Honavar and L. Uhr
        (Eds.) Artificial intelligence and neural networks: Steps toward principled integration. Boston: AcademicPress.

Bobrow, D. G. & Norman, D. A. (1975). Some principles of memory schemata. In D. G. Bobrow & A. M. Collins (Eds.),
        Representation and understanding: Studies in cognitive science. New York: Academic Press.

Chafe, W. (1994). Discourse, consciousness and time. Chicago: University of Chicago Press.

D'Andrade, R. G. (1995). The development of cognitive anthropology. Cambridge: Cambridge University Press.

Harkins, J. (1994). Bridging two worlds: Aboriginal English and cross-cultural understanding. St Lucia, Qld: University
        of Queensland Press.

Koriat, A., Goldsmith, M. & Pansky, A. (2000). Toward a psychology of memory accuracy. Annual Review of Psychology,
          Annual 2000 [on-line]. Available: Mailer-Daemon@epub.med.iacnet.com

Langacker, R. W. (1987). Foundations of cognitive grammar, Vol. 1: Theoretical Prerequisites. Stanford: Stanford
        University Press.

Langacker, R. W. (1991). Foundations of cognitive grammar, Vol. 2: Descriptive Application. Stanford: Stanford
        University Press.

Malcolm, I. G. (1977). Aboriginal primary school children and classroom speech: Some preliminary observations. In E.
        Brumby & E. Vaszolyi (Eds.), Language problems and Aboriginal education. Western Australia: Mt Lawley College
        of Advanced Education.

Malcolm, I. G. (1982). Communicative dysfunction in Aboriginal classrooms. In J. Sherwood (Ed.), Aboriginal Education:
        Issues and Innovations (pp. 153-172). Perth: Creative Research.

Malcolm, I. G. (1994a). Aboriginal English inside and outside the classroom. Australian Review of Applied Linguistics,
        17(2), 147-180.

Malcolm, I. G. (1994b). Discourse and discourse strategies in Australian Aboriginal English. World Englishes, 13(3),
        289-306.

Malcolm, I. G. (1995). Language and communication enhancement for Two-Way education. Report to the Department
        of Employment, Education and Training. Perth: Centre for Applied Language Research, Edith Cowan University.

Malcolm, I. (2000). The place of genres in Two-Way bidialectal education. Paper presented at the Fifth Annual Round
        Table of the Center for Applied Language and Literacy Research, Edith Cowan University, Claremont Conference
        Center, June-July, 2000.

Malcolm, I. & Rochecouste, J. (2000). Event and story schemas in Australian Aboriginal English discourse. English
        World-Wide, 22 (2), 95-123.

Malcolm, I. G., & Koscielecki, M. M. (1997). Aboriginality and English: Report to the Australian Research Council. Mt
        Lawley, WA: Centre for Applied Language Research, Edith Cowan University.

Malcolm, I. G., Haig, Y., Königsberg, P., Rochecouste, J., Collard, G., Hill, A., & Cahill, R. (1999). Two-way English:
        Towards more user-friendly education for speakers of Aboriginal English. Perth, WA: Education Department of
        Western Australia and Edith Cowan University.

Minsky, M. (1975). A framework for representing knowledge. In P. H. Winston (Ed.), The psychology of computer vision.
         New York: McGraw-Hill.

Nishida, H. (1999). A cognitive approach to intercultural communication based on schema theory. International Journal of
        Intercultural Relations, 23 (5), 753-777

Palmer, G. B. (1996). Toward a theory of cultural linguistics. Austin: University of Texas Press.

Rumelhart, D. E. (1980). Schemata: The building blocks of cognition. In R.Spiro, B. Bruce, and W. Brewer (Eds.)
        Theoretical issues in reading comprehension. Hillsdale, NJ: Lawrence Earlbaum.

Rumelhart, D.E., Smolensky, P., McClelland, J.L. & Hinton, G.E. (1986). Schemata and sequential thought processes in PDP
        models. In J. L. McClelland and D. E. Rumelhart (Eds.) Parallel Distributed Processing: Explorations in the
        Microstructure of Cognition Vol. 2: Psychological and Biological Models. MIT Press.

Schank, R.C. & Abelson, R. (1977). Scripts, plans, goals, and understanding.Hillsdale, NJ: Earlbaum Assoc.

Sharifian, F. & Samani, R. (1997). Hierarchical spreading of activation. In F.Sharifian (Ed.) Proceedings of the Conference
        on Language, Cognition, and Interpretation. Khorasgan: IAU Press. Available:
        http://www.angelfire.come/wy/fsh/act1.html

Strauss, C. & Quinn, N. (1997). A cognitive theory of cultural meaning. New York : Cambridge University Press.

Ungerer, F. & Schmid, H. J. (1996). An introduction to cognitive linguistics. London: Longman

* This is in fact known as the minmin light among Australian people. However, the association of the light with a spiritual presence in the car seems to be only held by Aboriginal Australians.