Pattern and Process in West Indian Archaeology Author(s): Irving Rouse Source: World Archaeology, Vol. 9, No. 1, Island Archaeology (Jun., 1977), pp. 1-11 Published by: Taylor & Francis, Ltd. Stable URL: http://www.jstor.org/stable/124458 Accessed: 13-02-2017 05:06 UTC JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://about.jstor.org/terms Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to World Archaeology This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
Pattern and process in West Indian archaeology
Mar 18, 2023
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Pattern and Process in West Indian ArchaeologyPattern and Process in West Indian Archaeology Author(s): Irving Rouse Source: World Archaeology, Vol. 9, No. 1, Island Archaeology (Jun., 1977), pp. 1-11 Published by: Taylor & Francis, Ltd. Stable URL: http://www.jstor.org/stable/124458 Accessed: 13-02-2017 05:06 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact [email protected].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to World Archaeology
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
Pattern and process in West Indian archaeology
Irving Rouse
It is customary to view the development of archaeological research in general terms - to speak, for example, of a current 'revolution' in the discipline and to distinguish between 'old' and 'new' archaeology. In fact, the discipline has developed differently from place to place, depending on the nature of the local situation and the interests of the local researchers, as this article will illustrate.
The purpose of the article is to discuss the use of the concepts of pattern and process in West Indian archaeology. By pattern is meant a configuration discernible in the archaeological remains and by process, the actions that have produced that pattern, whether carried out by natural or human agency. A pattern is an empirical observation and a process, an explanation of that observation; it tells us how the pattern came into existence.
Archaeologists disagree as to the relative importance of studying pattern and process. So-called 'normative' archaeologists concentrate upon patterns, in the belief that one should be as empirical as possible, whereas 'processual' archaeologists have become pre- occupied with processes; they are more interested in explaining archaeological remains than in understanding them (Binford I965; Flannery I967). It is the thesis of this article that West Indian archaeologists have succeeded precisely to the extent to which they have concerned themselves with both pattern and process.
Opinion also differs as to where the concepts of pattern and process fit into the pro- cedure of archaeological research. Some argue that we should proceed inductively by starting with patterns and postulating processes to explain the patterns (e.g. Rouse I972: I91-5). Others prefer to proceed deductively by hypothesizing a process and seeking empirical evidence in the form of a pattern - often called 'test implications' - that 'proves' the hypothesis (e.g. Watson, LeBlanc, and Redmond I97I). It will here be assumed that this, too, is a false dichotomy - that one should constantly keep both patterns and pro- cesses in mind and consider them in relation to each other. Without an awareness of
process, that is, of the ways in which archaeological remains can become patterned, one will not know what kinds of patterns to look for. Conversely, one must recognize a pattern before one can apply the principle of multiple working hypotheses by considering all processes that might have produced that pattern and selecting the combination of processes that is most likely to have done so (Chamberlain I965).
World Archaeology Volume 9 No. i Island Archaeology
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
2 Irving Rouse
Levels of archaeological interpretation
In its broadest sense, archaeological research aims to recover and to interpret both physical and cultural remains. Since the progress of research on West Indian physical remains, that is, on human skeletal material, is currently being reviewed elsewhere (Goodwin, MS.), I shall limit myself to cultural remains, that is, to the surviving products of human activity.
It will be assumed that the recovery and interpretation of cultural remains takes place on four logically successive levels, as shown in the first column of table io On the primary level, one seeks to learn as much as possible about the remains per se. One studies both the nature of the remains and their context, as indicated in the second column of the table.
On the secondary level, one turns one's attention to the cultural groups that produced the remains, that is, to peoples. First, one isolates the peoples and establishes their identity by determining which artefactual traits are diagnostic of each. In so doing, one concentrates on the most distinctive structures and artefacts and considers them prim- arily in terms of their technological and stylistic attributes. In other words, one's orientation is towards the building of the structures and the manufacture of the artefacts rather than towards their use.
Next, one studies the changes that have taken place in the artefactual assemblages laid down by each people, again approaching them from the standpoint of manufacture rather than use. The aim here is to determine how the diagnostic structures and artefacts developed among each people and from people to people.
On the secondary level, too, one studies the relationships of one's peoples to develop- mental groups of other kinds: the species and races distinguished by physical anthro- pologists and, if one is working in or near the historic period, the speech communities recognized by linguists and the peoples' own ethnic divisions. These other kinds of developmental groups are not shown in table i because, unlike cultural groups, they cannot be directly derived from cultural remains.
On the tertiary level, one turns from the peoples, their structures, and their manufac- tures to the peoples' behavioural systems, that is, to their cultures and social structures. The procedure is again two-fold; first one reconstructs the cultural and social systems and then one studies the changes in them (table i).
By a cultural system or culture is meant the sum total of the activities carried out by a people, including both the building of structures and the manufacture of artefacts, which
have already been covered on the secondary level of interpretation (Rouse I972, table 6). When reconstructing activities on the tertiary level, one examines the structures and artefacts from the standpoint of their use rather than their manufacture, that is, one focuses on their functional rather than their technological and stylistic attributes. One also examines the places in which each activity was performed and studies the by-products to be found there, such as food remains.
By a social structure is meant the total range of institutions through which a people carries out its activities, such as households, communities, governments, trade networks, and religious movements (Rouse 1972, table 6). These may be termed social groups, as opposed to the developmental groups studied on the previous level of interpretation. In
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
TABLE I
Nature of the Collection and identifi- Types, modes etc. Construction, manufacture Primary: remains cation or classification etc. inferences about
the remains Context of the Site survey and excavation Components, assemblages Occupation, deposition, remains etc. disturbance, decay etc.
Identities of the Classification and identifi- Complexes, styles, Temporal and spatial Secondary: peoples cation of the assemblages industries, phases etc. differentiation in tech- inferences about nology or style the peoples who produced the Changes in the Dating of the assemblages Patterns of change in Invention, diffusion, remains peoples' identities and study of their distri- types, modes, etc. migration etc.
bution
erarstructures social interactions inferences about
the peoples' Changes in the Dating of elements of the Variation in the Adaptation, demographic behavioral systems peoples' systems settlement patterns etc. elements change, dialectic conflict
etc.
Quaternary: Principles of Experimentation to test Test implications Co-variation in cultural generalizations about cultural and hypotheses derived from and social traits human development social change the principles
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
4 Irving Rouse
reconstructing them, one focuses upon the socially significant artefacts, such as status symbols and trade objects, and upon the attributes of those artefacts which are indicative of social groups, their composition, and their functioning. One also examines the places in which the institutions operated, seeking to learn as much as possible about their composition, the interaction among the individuals who belonged to them, and the interrelationships among the groups themselves.
Changes in cultural and social systems have to be studied with reference to the results obtained on the second level of interpretation. One turns again to the artefactual assem- blages, whose distribution in time and space has been established on the secondary level, and plots the distribution of the elements of culture and social structure against them. It is therefore not feasible to proceed to the tertiary level of interpretation until the secondary level has been well worked out; and the better the results that have been obtained on the secondary level, the better will be those on the tertiary level. This is an established principle in paleontology, where the assignment of organic remains to genera and species and the consequent determination of their evolutionary or developmental position is considered a prerequisite to studies of the functioning and adaptation of the organisms involved, but the principle is often overlooked in archaeology and among social anthropologists, who have tended to speculate about the course of human evolution without reference to the empirical evidence of archaeological assemblageso
Historic archaeologists do not need to pay so much attention to this principle as pre- historic archaeologists, because they have access to other kinds of evidence concerning behavioural systems such as documents, oral tradition, ethnographic accounts, and linguistic analyses. It is no accident that archaeological studies on the third level have achieved their greatest success when directed towards historic peoples or towards proto- historic peoples in regions blessed with rich bodies of data concerning historic peoples, which can be projected back into proto-history.
After having added knowledge of behavioural systems and their development to that concerning artefactual assemblages and their development, the archaeologist is finally in a position to make generalizations about human development by means of cross cultural study. He does so by comparing the artefactual assemblages, cultural systems, and social structures of all times and places in an effort to establish the existence of general trends
in human evolution, as opposed to the knowledge about the specific course of evolution in different parts of the world that has been acquired on the secondary and tertiary levels
of interpretation (Sahlins and Service I960: I2-44). This kind of research constitutes a quaternary level of interpretation. It has to come at
the end of the archaeological procedure because one cannot profitably carry it out without first having acquired sufficient knowledge of the peoples of the world, their artefactual assemblages, and their behavioural systems. As noted in Sahlins and Service (1960: 45-50), it involves different kinds of patterns and processes than those encountered in studying specific evolution. They have been termed nomothetic, as opposed to ideo- graphic, patterns and processes, because they refer to statistical tendencies rather than
specific events (e.g. Rouse I973). The third column in table i contains a list of the principal procedures used on each
level of interpretation. These procedures enable the archaeologist to discover the pattern- ing within the remains he studies, as exemplified in the fourth column of the table, and
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
Pattern and process in West Indian archaeology 5
to postulate processes that explain the patterns, as illustrated in the last column of the table. The procedures, patterns, and processes will now be discussed in the context of West Indian archaeology.
Primary interpretation in the West Indies
Recovery of remains and study of their nature as individual specimens began in the West Indies during the mid-nineteenth century. In 1847, Miguel Rodriguez-Ferrer started a collection in Cuba that included, among other artefacts, the 'lightning stones' found by the country people in their fields. He also recorded the existence of rectangular earth- works at the eastern end of the island (Harrington I921: 29-45).
In reporting on the discovery of such artefacts and structures, West Indian archaeo- logists have followed standard archaeological practice. Whenever possible, they have identified them in terms of the artefact categories current in our own civilization and especially in the discipline of archaeology (table i). Thus, Rodriguez-Ferrer noted that the 'lightning stones' were really celts. In a limited number of cases, it has also been possible to use the native categories, as these have survived in the historical writings about the aborigines of the West Indies. Identification of the earthworks as ball courts is an example.
Artefacts and structures that are foreign to our own civilization and are not mentioned in the historical sources concerning the Indians have had to be classified rather than identified (table i), that is, they have had to be assigned to new kinds of classes, based upon the similarities and differences in their attributes. Such classes are defined by listing their
diagnostic attributes and are named in terms of those attributes (Rouse I972, fig. 4). Some of the classes have been defined and named descriptively, as in the case of 'three-
pointed stones', which were presumably ceremonial objects. The pottery, on the other hand, has been grouped and defined in terms of its stylistic attributes. This emphasis upon descriptively or stylistically significant attributes contrasts with the emphasis upon functionally significant attributes in the case of the identified artefacts and structures.
Since potsherds comprise over 90% of the artefacts found in West Indian sites, the local archaeologists have paid particular attention to them. Two procedures are currently in use. One is to classify the sherds in terms of their individual features and the other to classify them as complete objects. In the former case, each of the resultant classes is defined in terms of a pattern of attributes of the kind of feature under study, and in the latter case, in terms of a pattern of the attributes of the sherds as a whole. The pattern of feature attributes is known as a mode and the pattern of sherd attributes, as a type (Rouse 1972: 48-50, 55-7).
The two patterns of attributes imply different processes of manufacture (table i). I have come to use the concept of mode because it is my impression that the West Indian potters built up their vessels feature by feature, using an additive procedure (Rouse 1965). The concept of type assumes instead that the potters made each vessel in accordance with a fully preconceived plan, as we ourselves do. It is a projection into the past of our own manufacturing processes, involving the use of blueprints and other records unavailable to the aborigines. But even if the concept of type were applicable to the aboriginal
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
6 Irving Rouse
situation, it would not be practicable in the West Indies, since we are forced by the nature of our finds to classify potsherds instead of complete vessels, to which the concept properly applies.
As in other parts of the world, the local archaeologists did not recognize the need to record the context of their remains until some time after they had begun to study the nature
of the remains. To be sure, many early workers did record the sites from which their artefacts came, but it was not until after the First World War that they began to note the
positions of the artefacts within sites. While digging the site of Ostiones in Puerto Rico, Adolfo de Hostos (I919: I4) observed that the red slip and modeled-incised decoration characteristic of the potsherds on the surface became rarer as he proceeded downwards through the refuse and eventually disappeared. He did not document this observation with measurements, however.
Since West Indian sites generally lack the remains of dwellings and other structures against which the positions of the artefacts can be plotted, and since few middens contain sharply defined layers which can be used for the purpose, documentation of the context of the remains had to await development of the procedure of excavating in terms of arbitrarily measured sections and levels. Froelich G. Rainey was the first to use this procedure, with my help (Rainey I94I).
The primary purpose of recording context is to be able to divide the site under excava- tion into occupations or components and thereby to isolate assemblages of artefacts, each laid down by a different people (table i). In effect, the components and assemblages constitute a pattern of distribution of the archaeological remains, which has to be explained in terms of the two pairs of processes listed in the last column of the table, occupation and deposition on the one hand and disturbance and decay on the other.
When archaeology began in the West Indies, it was assumed that, as in our own civiliz- ation, the aborigines lived in relatively permanent houses, cleaned them thoroughly, and deposited their refuse elsewhere in neat piles. Experience has shown that, probably because the Indians' dwellings were perishable, they moved them frequently from place to place. Moreover, they seem to have dropped their refuse haphazardly on the ground wherever they lived and to have trampled it into place, except when they were able to cast it down hill sides. As a result, the divisions between components tend to be horizon- tal rather than vertical and are often blurred by the haphazard nature of the process of deposition. Furthermore, a sharp vertical break between components does not ordinarily mean that a new group of people occupied the site, as the early writers assumed (e.g. Rainey I940). It is more likely to be a result of the process of shifting houses from one part of the site to another.
West Indian archaeologists have so far paid little attention to the processes of disturb- ance and decay and, as a result, have had little success distinguishing post moulds and other patterns indicative of the natives' houses and other structures. Egbert H. J. Boerstra (1974) is an exception; we need more excavations like his on Aruba.
Secondary interpretation in the West Indies
In I869, William M. Gabb (I872), an American geologist, found a sequence of pre- ceramic and ceramic layers in a cave on Samanat Bay in the Dominican Republic, but did
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
Pattern and process in West Indian archaeology 7
not attempt to define the peoples who had laid down the two layers. It was not until 1922-3 that Gudmund Hatt, a Danish archaeologist, initiated this kind of research while excavating in the Virgin Islands. He classified his sites into three groups, Krum Bay, Coral Bay-Longford, and Magens Bay-Salt River, and defined each group by noting its distinctive artefactual traits (Hatt I924). The sites in each group may be said to have been occupied by a different people, living in a different period, and possessing a different culture. Since we are here concerned with the concept of people (table i), I shall use that term in preference to the others.
The artefactual traits which Hatt used to define each of his peoples may be said to constitute a pattern. Such a pattern is comparable to the complexes, industries, or styles established elsewhere in the world. I shall here call it a complex, following my practice in
Rouse (I972: 84-7). Its basic use is to identify other assemblages laid down by the same people (table i).
Complexes presumably come into existence because local groups of…
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact [email protected].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
Taylor & Francis, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to World Archaeology
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
Pattern and process in West Indian archaeology
Irving Rouse
It is customary to view the development of archaeological research in general terms - to speak, for example, of a current 'revolution' in the discipline and to distinguish between 'old' and 'new' archaeology. In fact, the discipline has developed differently from place to place, depending on the nature of the local situation and the interests of the local researchers, as this article will illustrate.
The purpose of the article is to discuss the use of the concepts of pattern and process in West Indian archaeology. By pattern is meant a configuration discernible in the archaeological remains and by process, the actions that have produced that pattern, whether carried out by natural or human agency. A pattern is an empirical observation and a process, an explanation of that observation; it tells us how the pattern came into existence.
Archaeologists disagree as to the relative importance of studying pattern and process. So-called 'normative' archaeologists concentrate upon patterns, in the belief that one should be as empirical as possible, whereas 'processual' archaeologists have become pre- occupied with processes; they are more interested in explaining archaeological remains than in understanding them (Binford I965; Flannery I967). It is the thesis of this article that West Indian archaeologists have succeeded precisely to the extent to which they have concerned themselves with both pattern and process.
Opinion also differs as to where the concepts of pattern and process fit into the pro- cedure of archaeological research. Some argue that we should proceed inductively by starting with patterns and postulating processes to explain the patterns (e.g. Rouse I972: I91-5). Others prefer to proceed deductively by hypothesizing a process and seeking empirical evidence in the form of a pattern - often called 'test implications' - that 'proves' the hypothesis (e.g. Watson, LeBlanc, and Redmond I97I). It will here be assumed that this, too, is a false dichotomy - that one should constantly keep both patterns and pro- cesses in mind and consider them in relation to each other. Without an awareness of
process, that is, of the ways in which archaeological remains can become patterned, one will not know what kinds of patterns to look for. Conversely, one must recognize a pattern before one can apply the principle of multiple working hypotheses by considering all processes that might have produced that pattern and selecting the combination of processes that is most likely to have done so (Chamberlain I965).
World Archaeology Volume 9 No. i Island Archaeology
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
2 Irving Rouse
Levels of archaeological interpretation
In its broadest sense, archaeological research aims to recover and to interpret both physical and cultural remains. Since the progress of research on West Indian physical remains, that is, on human skeletal material, is currently being reviewed elsewhere (Goodwin, MS.), I shall limit myself to cultural remains, that is, to the surviving products of human activity.
It will be assumed that the recovery and interpretation of cultural remains takes place on four logically successive levels, as shown in the first column of table io On the primary level, one seeks to learn as much as possible about the remains per se. One studies both the nature of the remains and their context, as indicated in the second column of the table.
On the secondary level, one turns one's attention to the cultural groups that produced the remains, that is, to peoples. First, one isolates the peoples and establishes their identity by determining which artefactual traits are diagnostic of each. In so doing, one concentrates on the most distinctive structures and artefacts and considers them prim- arily in terms of their technological and stylistic attributes. In other words, one's orientation is towards the building of the structures and the manufacture of the artefacts rather than towards their use.
Next, one studies the changes that have taken place in the artefactual assemblages laid down by each people, again approaching them from the standpoint of manufacture rather than use. The aim here is to determine how the diagnostic structures and artefacts developed among each people and from people to people.
On the secondary level, too, one studies the relationships of one's peoples to develop- mental groups of other kinds: the species and races distinguished by physical anthro- pologists and, if one is working in or near the historic period, the speech communities recognized by linguists and the peoples' own ethnic divisions. These other kinds of developmental groups are not shown in table i because, unlike cultural groups, they cannot be directly derived from cultural remains.
On the tertiary level, one turns from the peoples, their structures, and their manufac- tures to the peoples' behavioural systems, that is, to their cultures and social structures. The procedure is again two-fold; first one reconstructs the cultural and social systems and then one studies the changes in them (table i).
By a cultural system or culture is meant the sum total of the activities carried out by a people, including both the building of structures and the manufacture of artefacts, which
have already been covered on the secondary level of interpretation (Rouse I972, table 6). When reconstructing activities on the tertiary level, one examines the structures and artefacts from the standpoint of their use rather than their manufacture, that is, one focuses on their functional rather than their technological and stylistic attributes. One also examines the places in which each activity was performed and studies the by-products to be found there, such as food remains.
By a social structure is meant the total range of institutions through which a people carries out its activities, such as households, communities, governments, trade networks, and religious movements (Rouse 1972, table 6). These may be termed social groups, as opposed to the developmental groups studied on the previous level of interpretation. In
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
TABLE I
Nature of the Collection and identifi- Types, modes etc. Construction, manufacture Primary: remains cation or classification etc. inferences about
the remains Context of the Site survey and excavation Components, assemblages Occupation, deposition, remains etc. disturbance, decay etc.
Identities of the Classification and identifi- Complexes, styles, Temporal and spatial Secondary: peoples cation of the assemblages industries, phases etc. differentiation in tech- inferences about nology or style the peoples who produced the Changes in the Dating of the assemblages Patterns of change in Invention, diffusion, remains peoples' identities and study of their distri- types, modes, etc. migration etc.
bution
erarstructures social interactions inferences about
the peoples' Changes in the Dating of elements of the Variation in the Adaptation, demographic behavioral systems peoples' systems settlement patterns etc. elements change, dialectic conflict
etc.
Quaternary: Principles of Experimentation to test Test implications Co-variation in cultural generalizations about cultural and hypotheses derived from and social traits human development social change the principles
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
4 Irving Rouse
reconstructing them, one focuses upon the socially significant artefacts, such as status symbols and trade objects, and upon the attributes of those artefacts which are indicative of social groups, their composition, and their functioning. One also examines the places in which the institutions operated, seeking to learn as much as possible about their composition, the interaction among the individuals who belonged to them, and the interrelationships among the groups themselves.
Changes in cultural and social systems have to be studied with reference to the results obtained on the second level of interpretation. One turns again to the artefactual assem- blages, whose distribution in time and space has been established on the secondary level, and plots the distribution of the elements of culture and social structure against them. It is therefore not feasible to proceed to the tertiary level of interpretation until the secondary level has been well worked out; and the better the results that have been obtained on the secondary level, the better will be those on the tertiary level. This is an established principle in paleontology, where the assignment of organic remains to genera and species and the consequent determination of their evolutionary or developmental position is considered a prerequisite to studies of the functioning and adaptation of the organisms involved, but the principle is often overlooked in archaeology and among social anthropologists, who have tended to speculate about the course of human evolution without reference to the empirical evidence of archaeological assemblageso
Historic archaeologists do not need to pay so much attention to this principle as pre- historic archaeologists, because they have access to other kinds of evidence concerning behavioural systems such as documents, oral tradition, ethnographic accounts, and linguistic analyses. It is no accident that archaeological studies on the third level have achieved their greatest success when directed towards historic peoples or towards proto- historic peoples in regions blessed with rich bodies of data concerning historic peoples, which can be projected back into proto-history.
After having added knowledge of behavioural systems and their development to that concerning artefactual assemblages and their development, the archaeologist is finally in a position to make generalizations about human development by means of cross cultural study. He does so by comparing the artefactual assemblages, cultural systems, and social structures of all times and places in an effort to establish the existence of general trends
in human evolution, as opposed to the knowledge about the specific course of evolution in different parts of the world that has been acquired on the secondary and tertiary levels
of interpretation (Sahlins and Service I960: I2-44). This kind of research constitutes a quaternary level of interpretation. It has to come at
the end of the archaeological procedure because one cannot profitably carry it out without first having acquired sufficient knowledge of the peoples of the world, their artefactual assemblages, and their behavioural systems. As noted in Sahlins and Service (1960: 45-50), it involves different kinds of patterns and processes than those encountered in studying specific evolution. They have been termed nomothetic, as opposed to ideo- graphic, patterns and processes, because they refer to statistical tendencies rather than
specific events (e.g. Rouse I973). The third column in table i contains a list of the principal procedures used on each
level of interpretation. These procedures enable the archaeologist to discover the pattern- ing within the remains he studies, as exemplified in the fourth column of the table, and
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
Pattern and process in West Indian archaeology 5
to postulate processes that explain the patterns, as illustrated in the last column of the table. The procedures, patterns, and processes will now be discussed in the context of West Indian archaeology.
Primary interpretation in the West Indies
Recovery of remains and study of their nature as individual specimens began in the West Indies during the mid-nineteenth century. In 1847, Miguel Rodriguez-Ferrer started a collection in Cuba that included, among other artefacts, the 'lightning stones' found by the country people in their fields. He also recorded the existence of rectangular earth- works at the eastern end of the island (Harrington I921: 29-45).
In reporting on the discovery of such artefacts and structures, West Indian archaeo- logists have followed standard archaeological practice. Whenever possible, they have identified them in terms of the artefact categories current in our own civilization and especially in the discipline of archaeology (table i). Thus, Rodriguez-Ferrer noted that the 'lightning stones' were really celts. In a limited number of cases, it has also been possible to use the native categories, as these have survived in the historical writings about the aborigines of the West Indies. Identification of the earthworks as ball courts is an example.
Artefacts and structures that are foreign to our own civilization and are not mentioned in the historical sources concerning the Indians have had to be classified rather than identified (table i), that is, they have had to be assigned to new kinds of classes, based upon the similarities and differences in their attributes. Such classes are defined by listing their
diagnostic attributes and are named in terms of those attributes (Rouse I972, fig. 4). Some of the classes have been defined and named descriptively, as in the case of 'three-
pointed stones', which were presumably ceremonial objects. The pottery, on the other hand, has been grouped and defined in terms of its stylistic attributes. This emphasis upon descriptively or stylistically significant attributes contrasts with the emphasis upon functionally significant attributes in the case of the identified artefacts and structures.
Since potsherds comprise over 90% of the artefacts found in West Indian sites, the local archaeologists have paid particular attention to them. Two procedures are currently in use. One is to classify the sherds in terms of their individual features and the other to classify them as complete objects. In the former case, each of the resultant classes is defined in terms of a pattern of attributes of the kind of feature under study, and in the latter case, in terms of a pattern of the attributes of the sherds as a whole. The pattern of feature attributes is known as a mode and the pattern of sherd attributes, as a type (Rouse 1972: 48-50, 55-7).
The two patterns of attributes imply different processes of manufacture (table i). I have come to use the concept of mode because it is my impression that the West Indian potters built up their vessels feature by feature, using an additive procedure (Rouse 1965). The concept of type assumes instead that the potters made each vessel in accordance with a fully preconceived plan, as we ourselves do. It is a projection into the past of our own manufacturing processes, involving the use of blueprints and other records unavailable to the aborigines. But even if the concept of type were applicable to the aboriginal
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
6 Irving Rouse
situation, it would not be practicable in the West Indies, since we are forced by the nature of our finds to classify potsherds instead of complete vessels, to which the concept properly applies.
As in other parts of the world, the local archaeologists did not recognize the need to record the context of their remains until some time after they had begun to study the nature
of the remains. To be sure, many early workers did record the sites from which their artefacts came, but it was not until after the First World War that they began to note the
positions of the artefacts within sites. While digging the site of Ostiones in Puerto Rico, Adolfo de Hostos (I919: I4) observed that the red slip and modeled-incised decoration characteristic of the potsherds on the surface became rarer as he proceeded downwards through the refuse and eventually disappeared. He did not document this observation with measurements, however.
Since West Indian sites generally lack the remains of dwellings and other structures against which the positions of the artefacts can be plotted, and since few middens contain sharply defined layers which can be used for the purpose, documentation of the context of the remains had to await development of the procedure of excavating in terms of arbitrarily measured sections and levels. Froelich G. Rainey was the first to use this procedure, with my help (Rainey I94I).
The primary purpose of recording context is to be able to divide the site under excava- tion into occupations or components and thereby to isolate assemblages of artefacts, each laid down by a different people (table i). In effect, the components and assemblages constitute a pattern of distribution of the archaeological remains, which has to be explained in terms of the two pairs of processes listed in the last column of the table, occupation and deposition on the one hand and disturbance and decay on the other.
When archaeology began in the West Indies, it was assumed that, as in our own civiliz- ation, the aborigines lived in relatively permanent houses, cleaned them thoroughly, and deposited their refuse elsewhere in neat piles. Experience has shown that, probably because the Indians' dwellings were perishable, they moved them frequently from place to place. Moreover, they seem to have dropped their refuse haphazardly on the ground wherever they lived and to have trampled it into place, except when they were able to cast it down hill sides. As a result, the divisions between components tend to be horizon- tal rather than vertical and are often blurred by the haphazard nature of the process of deposition. Furthermore, a sharp vertical break between components does not ordinarily mean that a new group of people occupied the site, as the early writers assumed (e.g. Rainey I940). It is more likely to be a result of the process of shifting houses from one part of the site to another.
West Indian archaeologists have so far paid little attention to the processes of disturb- ance and decay and, as a result, have had little success distinguishing post moulds and other patterns indicative of the natives' houses and other structures. Egbert H. J. Boerstra (1974) is an exception; we need more excavations like his on Aruba.
Secondary interpretation in the West Indies
In I869, William M. Gabb (I872), an American geologist, found a sequence of pre- ceramic and ceramic layers in a cave on Samanat Bay in the Dominican Republic, but did
This content downloaded from 125.22.40.140 on Mon, 13 Feb 2017 05:06:12 UTC All use subject to http://about.jstor.org/terms
Pattern and process in West Indian archaeology 7
not attempt to define the peoples who had laid down the two layers. It was not until 1922-3 that Gudmund Hatt, a Danish archaeologist, initiated this kind of research while excavating in the Virgin Islands. He classified his sites into three groups, Krum Bay, Coral Bay-Longford, and Magens Bay-Salt River, and defined each group by noting its distinctive artefactual traits (Hatt I924). The sites in each group may be said to have been occupied by a different people, living in a different period, and possessing a different culture. Since we are here concerned with the concept of people (table i), I shall use that term in preference to the others.
The artefactual traits which Hatt used to define each of his peoples may be said to constitute a pattern. Such a pattern is comparable to the complexes, industries, or styles established elsewhere in the world. I shall here call it a complex, following my practice in
Rouse (I972: 84-7). Its basic use is to identify other assemblages laid down by the same people (table i).
Complexes presumably come into existence because local groups of…
Related Documents
![m.a. Ancient Indian History, Culture & Archaeology(Semester System)[1]](https://static.cupdf.com/doc/110x72/5475e898b4af9fc80a8b5e7b/ma-ancient-indian-history-culture-archaeologysemester-system1.jpg)
