In a follow-up to Julianne’s previous post on scientific communication, I thought I’d describe a lecture I attended last week. I’ll try not to say anything overly controversial (though CV readers can be a tough crowd). The talk was by Felice Frankel, as part of the Santa Fe Institute public lecture series. The title was “More than Pretty Pictures: The Power of Images in Science”. Frankel is known for her scientific photographs. She creates beautiful images of a large range of physical systems (from water droplets to nanocrystals). She’s been responsible for quite a number of cover images for journals such as Science and Nature.
Frankel spent much of her lecture discussing her philosophy in taking scientific images. This consisted mostly of comments about the power of visualization, and ideas for how to make scientific visualization more effective. She emphasized that it’s highly nontrivial to produce an image which grabs you, while simultaneously informing you about the science it’s meant to represent. Many scientific images are uninspired. Or confusing. Often both. The lecture was sprinkled liberally with her work, much of which is quite arresting. For example:
This is an image of a ferrofluid. Frankel took seven small magnets, and placed them below a glass plate with the fluid above. She then added a bright yellow Post-It note below, yielding the vivid colors. It is this last step which completely transforms the photograph, and which your average scientist would have neglected. We have much to learn in how to present our results, both within the community, and to the world at large.
Images are indeed an essential component of science. They are visceral and physical, in a way that a table of numbers cannot hope to reproduce. They allow for what Frankel terms “visual thinking”: a direct and unmediated engagement with the world. This is particularly evident in astronomy. I would argue that the Hubble Space Telescope has generated many of the most beautiful images ever produced. And an appreciation of the science underlying the images only strengthens one’s admiration. Astronomy is peculiar in that a large portion of the field is fundamentally based on pretty pictures. (Okay, some of these pictures are run through variants of prisms to produce spectra, which aren’t quite as beautiful (at least to my, untrained, eye).) Julianne is our resident expert on taking and interpreting astronomical images; I’m told it’s a little more involved than pointing a digital camera and pushing the button.
What I found most surprising about Frankel’s lecture was her repeated insistence that she is not an artist, and that her photos are not to be considered art. As she put it: “This is why I am not an artist: I am deeply committed to maintaining the integrity of the science.” In her view, because she is constrained to reproduce the world as it is, she is not allowed the free rein of an artist. Her focus is on communicating science as effectively as possible: education rather than aesthetics, meaning rather than art. I find this argument somewhat disappointing. Her most effective images are certainly art; in fact, a number of museums have added her photographs to their collections. And her ability to produce these images, without the liberty of composing unphysical scenarios, or the liberal application of photoshop, does not detract from her talents. If anything, the restricted domain in which she works emphasizes her abilities. Although the sonnet is a severely constrained form of expression, I don’t see anyone arguing that Shakespeare’s contributions don’t qualify as art.
One side-note which Frankel briefly touched upon was the issue of “true” or “accurate” representation in science. While Frankel makes an effort to maintain the essential integrity of her images, most Hubble images are somewhat enhanced (false-color). This means that, were you to manage to stick your head into the focal plane of the Hubble telescope (the fact that it’s hundreds of miles above the surface of the Earth notwithstanding), the image you would see with your eyes would look completely different from the postcards we’re all familiar with. Scientists have taken liberties with the color palette and contrast in producing the images. Often the frequencies of the light in astronomical images are well outside human experience. The human eye is a particular sensor, and there’s no reason that it “sees” the universe in a way that’s in any sense profound. For example, we don’t see infrared. If we did, a hot pan on the stove would glow as a warning, and all those times I have dropped spaghetti sauce all over the floor would have been avoided. We don’t see x-rays either (superman presumably does; but in his case his eyes must not only be sensitive to x-rays, but also emit them in the first place, since the Sun isn’t bright enough in x-rays to give him good images on Earth). There are interesting astronomical sources of light at essentially all frequencies we’ve cared to observe, and so we generate images in a tremendous range of wavelength bands. Furthermore, by playing with the contrast and color scale, we can highlight various features and structures in the images; perhaps we’d like to “see” star forming regions, or shocked gas, or interstellar dust. As a happy byproduct, we also make the images visually stunning. It’s probably not entirely happenstance that images which emphasize interesting science also happen to be more beautiful. Although you would never see the identical scene with your naked eye at a telescope, the images are no less physical or instructive. They represent good science and good aesthetics. What’s not to love?