Science of Culture

Participants in the Convergent Technologies (NBIC) conference recommended a new scientific initiative, analogous to the Human Genome Project that charted the human genetic code, which they called the Human Cognome Project — unraveling the secrets of the human cognitive genome. Any attempt to solve the riddles of the human mind will have to be far more than an exercise in brain neurology; most importantly, it will have to attack the mysteries of the cultural genome.

One major benefit of a program in memetics would be to better understand culture as an evolutionary process in its own context, whether as a Darwinian, Lamarckian, or as yet unknown system (Boyd and Richerson 1985). The knowledge gained could create a framework for a scientific rebirth in social and cultural domains. While opinions vary, it would not be too harsh to suggest that several of the social sciences seem to have stalled, some of them achieving very little progress in recent decades. The same thing occasionally happens in physical sciences. For example, planetary astronomy had practically stalled in the two or three decades prior to the launch of the first interplanetary space probes. Similarly, cancer research has achieved progress only very slowly over the past century, but the Human Genome Project offers new hope of breakthroughs. Memetic science could provide just the intellectual boost and potent research methodology needed by such diverse fields as Anthropology, Political Science, and Sociology.

Development of new theories and methods will require cooperation between hundreds of scientists in perhaps a dozen fields, so here with our limited perspectives we can suggest only a few of the possibilities. Perhaps there are a number of common features of natural codes, including both cultural and biological codes:

• The "independence"feature: Natural code elements tend to have arbitrary meaning (C.S. Peirce's symbols, as opposed to icons or indices) facilitating abstraction and reuse.

• The "combinatorial advantage" feature: The number of potential representations is much larger in combinations of elements than in one-to-one element coding — perhaps because evolutionary selection favors representational richness available by combination sets.

• The self-regulation of natural codes: Dependency upon a code results in a constraint for new input to be interpreted in terms of the code; change is thereby limited to evolution of the code over time.

Work on applying language modeling to genomic sequences at Carnegie Mellon University has suggested that genomes differentiate species by having distributions that include rare occurrences and where such rare occurrences can often be species-unique. This work suggests that some species-unique sequences have an unusual generative power, such as those playing an important role in fold initiation of proteins. Perhaps cultural codes also contain some rare occurrences that serve to differentiate cultures and are heavily associative, or generative, within the culture.

The study of cultural codes, such as suggested here, has not proceeded as rapidly as other fields such as bioinformatics. Perhaps there are reasons of politics and objectivity that have lowered the expectation of resources available for doing such research. Cultural codes may be easier and more politically feasible to study in the short-run in culturally primitive groups or other large-brained species. Bottlenose dolphins, for example, participate in fluid, short-term social associations, and their vocal plasticity as well as their behavior appears to be related to their fission/fusion social structure (Reiss et al. 1997). Perhaps dolphins' fluid social groups provide external cognitive representations (perhaps via "mirror neurons") in a manner similar to the totems of primitive human cultural groups.

Several systematic research methodologies need to be developed. One breakthrough that seems within reach would be the memetic equivalent of the Linnean system for classifying species, genera, and other kinds of biological clades. In recent years, information science has developed a range of techniques, such as latent semantic analysis and semantic concept space technology (Harum et al. 1996). United with cognitive science, these methods should go a long way to identifying the structure of the cultural genome and the mechanisms by which it changes or sustains itself. Through the development of memetic science, we will want to look to genetics for inspiration and selectively import both theories and methods from biology when appropriate.

The scientific study of culture is both possible and pregnant with knowledge of human behavior. Thus, it deserves to be given more resources, especially in light of current events. These events include not only the terrorism of September 11, 2001, but also the dot-com crash and the failure of nations as diverse as Argentina, Indonesia, and Japan to sustain their economic development. Memetic science could help us deal with challenges to American cultural supremacy, discover the products and services that will really make the information economy profitable, and identify the forms of social institutions most conducive to social and economic progress.

0 0

Post a comment