A digital 'New Wild'
The scientific imagination is now profoundly digital in a way that owes much to Turing, Von Nuemann and the first days of scientific computing. The growing science of complexity and 'emergence' in natural systems merely adds to this sense of a new lens through which to view nature. We can now begin to simulate and analyse patterns in nature with the concept of ‘algorithms’ and ‘systems’ as something intrinsic to nature itself (see Stephen Wolfram's 'A New Kind of Science' or Mitchell Waldrop's book 'Complexity'). It is only through computers that we are able to view nature this way, because only computers can ‘capture’ the complexity that we try to describe with elementary or complex programs. For a gentle introduction to the concept of complexity see the article 'Complexity 101' in this blog. This website will showcase scientists who adopt this view of complexity and who also deal with the challenges that it produces. It will also be argued, however, that with the advantages of this new view of complexity in nature also come potential costs. These costs are the subject of this article and may shed light on the limits and challenges of the ‘complex’ view of nature, and some of the features of the ‘New Wild’ that this website project is about.
One aspect of this project is concerned with the idea that the risk of modern complexity is that we use digital resources to analyse and solve problems which may merely exchange one form of complexity for another. This is one feature of the ‘New Wild’. It is argued that this pattern may also be seen in scientific research that seeks to extend our knowledge of nature using huge amounts of digital resources.
The use of digital resources to tackle scientific questions about nature is increasingly seen as the answer to penetrating all cases of natural complexity and also now shapes nature as complex. In part, this is due to the recent history of great successes in scientific computing. Research in physics at CERN was arguably the beginning of a new age of large scale scientific computing which also saw the birth of the World Wide Web. The race to map the human genome was also an important landmark in scientific computing. More recent projects include the use of huge computer simulations in climate change science, or the ‘Brain Initiatives’ in the US and the European Union. New approaches also include data-mining of 'big data' (using machines to find patterns in huge data sets) which has had great success in genetic science. Yet, even the successes of complex digital solutions to scientific problems can create huge problems for science, for similar reasons that it is argued elsewhere on this website that it has created problems for society.
Digital science may actually remove us further from more direct contact and a more concise understanding of nature. In the ‘systems’ view, relatively simple algorithmic structures were once thought to potentially capture awesome apparent complexity in a simpler underlying process. This, in a sense, offered the opportunity to bring us ‘closer’ to nature by simplifying the complex. However, the biggest scientific results to date stem from the use of scientific computing to carry out huge simulations or analyse huge data sets with massive amounts of computer resources.
Despite significant successes, these uses of massive computer resources are sometimes unable to resolve lingering uncertainties about how results should be interpreted or even how problems of understanding should be stated or communicated by human beings. These methods also therefore create problems for the publishing and sharing of scientific data. It may also not always be clear what it now means to understand a natural phenomenon via computer simulation of that phenomenon, or via machine learning algorithms.
The lack of an understanding of natural complexity may therefore be merely exchanged and fused with a lack of complete understanding of the technological (digital) complexity used to analyse it. There is therefore, a possibility that the huge availability of digital resources may eventually cause scientists to denigrate what it means to understand something, scientifically. This is touched on by the mathematician and astronomer, John D Barrow in his book: ‘Impossibility: The Science of Limits and the Limits of Science’ where he argues that the ability to create complex simulations of nature may not really be ‘understandable’ in the way that a reductive, traditional, scientific law, is. He says, (albeit, rather whimsically): ‘A full simulation of a complicated natural phenomenon would involve a program of complexity approaching that of the thing being studied. It is like having a full-scale map, as large as the territory it describes: extraordinarily accurate; but not so useful, and awfully tricky to fold up’. Indeed, in many cases, the extensive use of computer simulation to model nature, or data mining to find patterns in data, have been met with varying degrees of scepticism by other scientists. This skepticism stems, in part, from the perceived complexity of both the technology and the aspects of nature now being studied by that technology.
What is clear, is that scientific computing offers great potential in providing a means of tackling huge, previously hidden, natural complexity and has a track record of success. In addition, the modern view of nature as complex opens up a whole new way of viewing natural processes as algorithm-like structures which can simplify apparent complexity. However, this power to view nature in all its complexity doesn't come without a price: The risk of misinterpretation and degradation of what we mean when we say we understand nature is part of that risk. Scientists apparently succeed in the digital replication of natural systems via simulation or the discovery of patterns in nature due to data mining and machine learning. However, the digital complexity we have used to 'solve' the 'understanding nature' problem may also be at the edges of effective human understanding. If this is the case another profound form of human mastery, knowledge of natural laws, begins to disappear from view. The replacing of direct scientific experimental contact with nature, with the new complexities of digital resources is therefore argued to be a central component of 'The New Wild': It signals a new limit on our understanding, signalling both profound changes in our conception of nature and ultimately, our relationship to nature, too.
Barrow, John D. (1998) 'Impossibility: The Limits of Science and the Science of Limits'. Oxford University Press.
Langton, Christopher G. 'Artificial life: An overview'. Mit Press, 1997.
Waldrop, Mitchell M. 'Complexity: The emerging science at the edge of order and chaos'. Simon and Schuster, 1993.
Wolfram, Stephen. 'A new kind of science. Vol. 5'. Champaign: Wolfram media, 2002.