What is the ‘process’ in cultural process and in processual archaeology? (d)

(being continued from 12/03/19)

SHIFTING MEANINGS OF CULTURAL PROCESS
I suspect that what made the concept of cultural process difficult to discuss in the 1970s
and 1980s was that it was a ‘vogue word’ – a word that was used to show that the user
had acquired the term and all the supposed intellectual accessories that went with it (Service, 1969). M. Johnson (1999: 188) correctly suggested that conceptual ambiguity
was exacerbated because the term process could be ‘used by an author in one sense and read in quite another’. This ambiguity allowed an unremarked shift in the meaning of the term over time.
Binford (1968b; see also Binford and Binford, 1968b) not only recognized the difference between the category of processes comprising the synchronic operation family and
the diachronic evolution family, but he also noted that the archaeological record had
been created by human behaviors, among other processes, and that the creation processes comprised a different category (Figure 2). He argued that an archaeologist must distinguish between explaining the archaeological record – a modern static phenomenon –and an extinct cultural system – a past dynamic phenomenon. A processual archaeologist seeks to understand ‘processual relationships among various classes of material items in the dynamics of cultural systems’, but such understanding can only come after one has explained observations made on the archaeological record [which] necessarily involves coping with problems of process. We attempt to explain similarities and differences in archaeological remains in terms of the functioning of material items in a cultural system and the processual features of the operation or evolution of the cultural systems responsible for the varied artifact forms, associations, and distributions observable in the ground. (Binford, 1968b: 273)
In the last quoted sentence, the wording before the second ‘and’ refers to behavioral
processes that create the archaeological record; the words after that refer to the dynamic cultural processes (both synchronic operational and diachronic evolutionary) at work when the culture existed.

The distinction is made by an archaeologist interested in contributing to anthropological theory (Binford, 1962). It, along with Schiffer’s (1976) more detailed discussion,ultimately led to refocusing the processual approach not on what might be thought of as the ultimate goal of identifying category 2 processes – those involving the operation and evolution of dynamic cultural systems – but on the more archaeologically proximate goal of category 1 processes – what became known as processes that result in the formation of the archaeological record (Schiffer, 1987). (I use 1 and 2 merely to signify order of analysis.) By the middle 1980s, for example, the text of an edited book entitled Structure and Process in Southeastern Archaeology, comprising 14 chapters authored by 16 archaeologists (Dickens and Ward, 1985), hardly mentions category 2 (cultural) processes and instead focuses nearly exclusively on category 1 (formation) processes.
Schiffer (1988) highlighted the distinctiveness of the two categories when he distinguished between ‘reconstruction theory’ (concerning category 1 processes) and ‘social theory’ (concerning category 2 processes). In her introduction to the recent volume Processual Archaeology, A.L. Johnson (2004) suggests that processes involve how a cultural system operates and also how the archaeological record is formed.
Additional evidence that formation processes usurped the priority of cultural
processes is found in recent perspectives on Wauchope’s dilemma. Keegan (1991)
suggests that at least some processes of interest are the behavioral processes that created the archaeological record, and others are the cultural processes or dynamics of an existing culture. Following Terrell (1986), Keegan (1991: 186–7) indicates that ‘cultural processes are the unfolding patterns of variability through time in conformity with nomothetic principles’. Terrell (1986) seems to have skirted Wauchope’s dilemma by suggesting that repetitive archaeological patterns were the results of predictable types of patterned human behavior. Echoing Kroeber (1948), Cunningham (2003: 391) indicates that ‘causal processes interact and combine in the creation of material patterns’.
This results in three things. First, a particular pattern ‘can be created by entirely different sets of causal processes’ (2003: 392); this is equifinality (Lyman, 2004). Second, analytical reconfiguration of processes allows one to ‘explain behavior that may differ substantially from any modern situation’ (Cunningham, 2003: 394). And third, different
processes and combinations thereof create different patterns in the material record
(2003: 395). Identifying which processes and combinations thereof create which
patterns is not only the goal of formation process studies (Schiffer, 1987) and middle
range research (Cunningham, 2003), but it is, seemingly, now of importance equal to
that of identifying the cultural, social, and behavioral processes that processual archaeologists originally sought.

DISCUSSIONS OF CULTURAL PROCESS BY PROCESSUAL ARCHAEOLOGISTS
Eisenberg (1971) argued that Deetz’s (1965) analysis exemplifies processual archaeology
because it adopts the view of a culture as a system and seeks to understand the role of
kinds of artifacts (in Deetz’s case, ceramic decorations) in a cultural system. It was in
fact a systems-theory perspective that produced the two most detailed discussions of
cultural processes by archaeologists of which I am aware. One was by David Clarke; the
other was by Frank Hole and Robert Heizer.
Clarke (1968: 22) wrote that the ‘primary processes are those of inevitable variation,
multilinear development, invention, diffusion and cultural selection. Combined in
many permutations and circumstances these processes give rise to such complex processes as acculturation, and cultural growth, decay and disintegration’. He explicitly
defined a process as ‘a vector which describes a series of states of an entity or system
undergoing continuous change in space or time’ (Clarke, 1968: 42, 668). Clarke (1968:
43) believed that a ‘general system model . . . should be representative of cultural
processes at several levels within a sociocultural unit’. He sought ‘a model for archaeological processes – archaeological entities changing as special kinds of dynamic systems,susceptible to analysis in terms of general systems theory’ (1968: 72). He indicated that ‘We should not expect the processes that operate upon cultures or culture groups to be the same as those that operate upon artifact attributes, although since the former entities are compounds of the latter elements we might expect the processes appropriate to higher entities to integrate the simpler processes as well’ (1968: 409). The first part of the immediately preceding sentence echoes Wauchope’s dilemma, and the last part concerns the magnitude and scale of processes.
Clarke (1968) lists various processes and provides a detailed and relatively lengthy
statement on category 2 (cultural, behavioral, social) processes that he thought operated on or within most sociocultural entities. He specified three ‘general processes’
ontogeny, migration, interaction – and then suggested that each was manifest in various
ways by development of variants, decrease of variety, increase of variety, and transformation of variety of cultural elements (Clarke, 1968: 409–10). The problem, Clarke thought, was to derive a ‘nested hierarchy of socio-archaeological processes’ – processes that linked sociocultural change to change in artifacts of whatever scale (Clarke, 1968:411); this is an effort to resolve Wauchope’s dilemma. Each process Clarke listed is a general kind that includes more specific kinds of processes. Many of his processes fall within the diachronic evolutionary family.
Changes in Hole and Heizer’s multi-edition introductory textbook capture the
growing importance of culture processes in archaeology. There is minimal mention of
culture processes in the first two editions (Hole and Heizer, 1965, 1969), but the 1969
edition contains a discussion of the importance of systems theory to understanding
cultural dynamics. The third edition (Hole and Heizer, 1973: 439) discusses cultural
processes explicitly.
The term ‘process’ or ‘processes’ crops up frequently in the writings of scientific
archeology, and it is also used in history, in manufacturing, and in analysis. As we
understand the term colloquially it refers to the sequential set of operations that lead
from A to B . . . [Given examples in history, manufacturing and research one] can
readily see that process means two quite different things. First, it may refer to a
sequence of events. Second, it may refer to the causes of the sequence of events. In
both meanings, process is conceptually linked with the states or conditions of the
things under observation at different times. As process is used in archeology, it refers
to an analysis of the factors that cause changes in state.
The authors provide the same discussion in an abridged version of their book (Hole and
Heizer, 1977: 358), where they also define ‘process’ in the glossary as ‘the operation of
factors that result in a change of culture’ (1977: 387). Note that Hole and Heizer
indicate that a process can be a simple description of a sequence of events, or it can
refer to cause(s) of that sequence. Given processual archaeology’s hopes to explain the archaeological record rather than just describe it (Lyman and O’Brien, 2004), a reasonable inference is that processualists sought to identify causal processes that operated prehistorically.
One of the alleged benefits of archaeologists adopting systems theory was that ‘questions phrased in terms of [systems] concepts direct our attention away from institutions and events and toward processes, away from efforts to discover the first appearance of particular cultural practices and toward efforts to understand their gradual evolution,and away from constructions of these events that are relatively hard to define in terms of archaeological observations toward ones that are more sensitive to the data with which we deal’ (Plog, 1975: 215). Plog is unclear, but I suspect he hoped to identify dynamic cause(s) rather than describe static events in temporal terms. Thus perhaps Plog was concerned with how a cultural system operates. Salmon (1978: 175), after all, pointed out that ‘anthropologists were engaged in analyzing social and cultural systems long before the advent of modern systems theory’ (see for example Kluckhohn’s [1951] discussion of Linton [1936]). This is particularly evident if one is aware of the structural–functional approaches in anthropology early in the 20th century, and also of the typical definition of a system as the relationships (mechanical, structural, functional)between entities comprising the system (Hill, 1977; Maruyama, 1963; Plog, 1975;Salmon, 1978). Systems theory seems to be preadapted to studying the dynamic operation (static state of being) of a cultural system.
When Clarke wrote his magnum opus, he modeled his recommendations for archaeological research on systems theory. But he did so with the following explicit and emphatic caution:
It would be all too easy to take systems theory as our model for archaeological [that
is, sociocultural] processes and the cultural entities that generate them, without isolating precisely the kind of system these entities represent. This would simply extend
systems theory and its terminology as yet another vague analogy of no practical
potential. (Clarke, 1968: 39)
Because in Clarke’s (1968: 39) view, anthropologists were ‘only just beginning to analyze
social systems in [systems theory] terms’, he devoted the majority of the nearly 700 pages of Analytical Archaeology to building a model of culture, including artifacts, as a system.
He used systems theory concepts and terms such as ‘feedback’ and ‘homeostasis’ in his
modeling efforts, but his cultural processes were not categorized by him as the generic
deviation counteracting ones of the first cybernetics meant to study stasis nor the deviation amplifying ones of the second cybernetics meant to study change (Maruyama,
1963). Rather, they were ‘technocomplex repatterning’, ‘culture group repatterning’,
‘acculturation’, ‘diffusion’, ‘invention’, and the like (Clarke, 1968: 410–12). Flannery
(1968, 1972), on the other hand, categorized the cultural processes he identified (seasonality, scheduling, centralization, segregation) as either one or the other of these two general categories (deviation counteracting or amplifying), and he, like Clarke, identified and named specifically cultural processes.
A significant influence on processual archaeologists’ focus on cultural processes was
Maruyama’s (1963) discussion of the second cybernetics, or ‘deviation-amplifying
mutual causal processes’ (see Flannery, 1968, 1972; Hill, 1977; Plog, 1975 for an introduction). In his 16-page article, Maruyama used the term ‘process’ 40 times, or 2.5
times per page. What Maruyama means by the term is never explicit. I suggest that he
meant several things, including dynamic causes, relationships (mechanical, functional)
between variables or entities, and influences of one variable or entity on another. Perhaps because it is unclear whether he meant static mechanical relations or causally dynamic ones, to this day some authors define a cultural process such as ‘centralization’ as ‘the degree of linkage between the various subsystems and the highest-order controls of a cultural system’ (Spencer, 1997: 215). This definition does not identify a dynamic cause though it does imply some kind of mechanical or functional relationship between phenomena. The name of a process has been applied to both a dynamic cause and its result. This seems to be the way that Maruyama (1963) used the term process, and his use likely influenced archaeologists.

(to be continued)

R. Lee Lyman
University of Missouri-Columbia, USA