Painting by Numbers: Image, Data and the Mathematical Sublime in Late Twentieth Century Astrophysics, Max Planck Institute for the History of Science, Berlin (May 2006)
Very large data sets are ‘collections’ too: in an increasingly digital world, we need to understand their materiality as continuous with other forms of material culture. We need to understand how they are generated, how they are analysed, how they produce knowledge, and what this means in epistemological terms.
Shortly after the completion of my NESTA residency at the Institute of Astronomy, University of Cambridge, I was invited to present my research findings at the Max Planck Institute for the History of Science in Berlin.
Under the direction of Professor Lorraine Daston, the ‘History of Scientific Observation’ project had just got underway, and I chose to make a contribution to this project with an extensive paper about numerical databases in recent astrophysical research. In particular, the paper concerns the coming into being, use and impact of the Automatic Plate Measuring Machine which for thirty years had been at the heart of data generation in UK astronomy.
This pivotal instrument was designed by Ed Kibblewhite on the cusp of the 1970s and operated by Mike Irwin at the Institute of Astronomy until 2005, when it was finally decommissioned. The photograph you see above, taken by IoA Graphics Officer Amanda Smith, shows it leaving the building that was constructed to house it.
The main function of the APM was to scan very high quality all-sky survey photographic plates of the near universe and turn the numerical data generated into coherent, searchable databases. Its use precedes the wide use in telescopes of particle detecting CCDs (charge couple devices) which deliver a constant stream of numerical data. As such, it was the machine that produced an entire algorithmic lexicon for understanding the universe.
Here is an abstract of the paper, which I am currently preparing for publication:
What does “observation” mean in a digital age and how is this related to its origins in eras when visual culture was more physically tangible? Beginning with the material culture of astronomical photography and extending into current computational astrophysics, this paper traces the intertwined evolutions of data and image in astronomical practice. I will argue that, far from there existing a philosophical or methodological split between those practitioners who use images and those who use logics as ’observational’ tools, in astronomy image and logic are synonymous, collapsed into each other.
I will explore this phenomenon in part through its origin in traditional observational practices of technical and computational data extraction from photographs of the sky over a hundred year period, showing how this ’tribal memory’ affects not only contemporary astronomers’ relationship with avalanches of post-digital data, but also in turn culturally informs the production of present-day images synthesized from the accumulated data itself. The paper pays close attention particularly to the 1970s, a period in which overlap between the material culture of ’sky survey’ photographs, the design of automatic measuring devices and the rapid evolution of computer power — all functioning at the very limit of their capacities — created a nexus of image-data systems which enshrined the mobile equivalence between the two.
This evolving relationship from image to data and back again to image via scanners, computers and display technologies is a very important one for science in general and for culture at large in the last 40 years. In astronomy, the build-up of large data sets in what has been a supremely visual science of light extends the very notion of what the adjective ‘visual’ means and thus it is a good test case to examine these cultural changes.
Starting with an overview of recent historical, sociological and art historical attention paid to late 20th century astrophysics, I show different approaches to the visual culture of astronomy and to notions of aesthetics. I conclusion, I propose a mathematical sublime at the core of the production of photo-illusory visualisations of the universe produced for public consumption, and suggest that the feature missing from a clear understanding of all image-making in contemporary science is the still undeveloped context of a logic of aesthetics.
It was to be the first of several visits to the Max Planck Institute for the History of Science as a guest researcher. On that occasion in 2006 it was fantastic to be able to compare notes across astronomical image and data with photo historian Dr Kelley Wilder (now running the Photographic History Research Centre at De Montfort University) and historian of 19th century physics and astronomy, Dr Charlotte Bigg (now a senior researcher at the Centre Alexandre Koyré in Paris), both of whom were at the MPIWG at that time.
I have also more recently been a guest researcher again at the MPIWG in March of 2012, developing an exhibition project and giving the Institute Colloquium on 21 March 2012 — about the very different subject of natural history museums.
[Image References: the decommissioning of the Automatic Plate Measuring Machine (Amanda Smith, 2005); black and white photo of the prototype automatic plate measuring machine by Ed Kibblewhite]