Walking Downtown Davis (B-G/1-4)

Hello friends and family fans of Horiátiki!

We’ve been silent for a while now – and no, unfortunately, we’re not back in Greece. I will remind you, however, that the exact translation of horiátiki is “a little something from the village” – it doesn’t necessarily mean a Greek village. And so, after a long hiatus, I’m resurrecting Horiátiki (the blog) to bring you some thoughts and images of our very own village of Davis, CA.

"What’s the occasion?" you might ask. “You’ve lived in Davis for almost 5 years now and haven’t felt compelled to take pictures or write about the place before, so why now?"

Good questions.

The short answer is that I was inspired by (and, yes, required to complete) a culminating project for one of the very last classes that I will ever take in all of graduate school - “image-based field research.” In this course, we talked about how social scientists (which apparently I am - at least for the time being) can and do use images like still photographs and video to collect information about the world, investigate research questions, and communicate and share ideas with others.

In the project that I'm sharing with you here, I explore two different ways of collecting photographic data about downtown Davis. I hope you enjoy the discussion and more importantly, I hope you enjoy getting a glimpse of the place where we’ve been living for the last few years. I certainly had fun getting to know the place in a different way through this project.

The art and science of image-making in social science research

In a 1976 interview printed in The CoEvolution Quarterly, renowned anthropologists Margaret Mead and Gregory Bateson argue (in the way that only amiable, formerly married people can) about whether image-based social research should be considered an “art” or a “science.” Bateson asserts that “the photographic record should be an art form” in that the researcher should be free to capture photographs and footage of whatever happens to be interesting or relevant in a particular social setting. He insists that the researcher should have "control of a camera” and rails about the practice of “putting a dead camera on top of a bloody tripod.”

Mead, on the other hand, dismisses Bateson’s method of “jumping around taking pictures.” “Why the hell should it be art?” she asks. She argues, instead, for a more systematic, scientific approach to using photography and film/video in social research and sees nothing wrong with using a camera on a tripod. “I think it’s very important, if you’re going to be scientific about behavior, to give other people access to the material as comparable as possible to the access you had. You don’t then, alter the material.” Bateson’s method of “leaping around,” she contends, alters the material in such a way that it doesn’t get at “what is happening” in a particular social setting; instead, it gets at what he thinks is happening.

Of course every social science researcher, using visual methods or not, needs to contend with the sticky proposition of claiming to represent social reality. Our interpretations of what is happening in the world around us, no matter how carefully and systematically we collect and analyze our data (either on our own or in collaboration with “subjects”), can never wholly represent the “truth” of any particular situation or bypass our own subjective experience. At the same time, those of us who believe that the endeavor of social research is a worthwhile pursuit, despite it's obvious limitations, know that it is only by being carefully systematic about our methods that we can hope to instill confidence in our representations and interpretations.

The challenge of negotiating the balance between "art" and "science" assumes a particularly blatant form for researchers who make and use images in their work, namely: While the power of imagery to communicate emotions, evoke senses of place, and tell stories depends critically on the myriad decisions that a photographer or videographer/filmmaker makes in capturing visually-arresting scenes, these very "artistic" decisions can compromise the scientific objectivity required of legitimate social researchers by their research communities.

So, how does one balance the "art" and "science" of image-collection in social science research? Is it possible to acquire visual data in a systematically non-biased way, true to Mead's ideal, while simultaneously accommodating Bateson's more artistic "leaping around" methodology as a means of capturing visually-arresting images of a place or situation?

In this little project, I explore two different methods for collecting photographic data about “what is happening” in Davis. Both methods are attempts to explore a certain aspect of the town, specifically the physical landscape of downtown Davis between B-G and 1-4 streets, in a systematic way.

As you take a look at the photographs, please be thinking about the following questions:

Which method do you think more accurately represents downtown Davis?
Which method produces the most informative images? (i.e. If someone were presented these images 100 years from now without knowing anything about the town or society, which would give them a better idea of what life is like in this place?)
Which method produces the more evocative or visually-interesting images?

Method 1: Making pictures with a "shooting script” and capturing visually-arresting images

Various image-based researchers employ “shooting scripts” as a way to systematize data collection and avoid bias and error when taking pictures of places and social phenomena (Wagner, 1979; Schwartz, 1989; Suchar, 1997). Shooting scripts are simply lists of research topics or questions that can be examined via photographic information (Suchar, 1997).

Suchar (1997) notes that the shooting scripts employed by the Farm Security Administration’s (FSA) photographic team in the 1930’s and 1940’s (including Dorothea Lange, Walker Evans and Arthur Rothstein) deserve at least part of the credit for making the FSA photographic study one of the most important visual studies ever undertaken.




(Arthur Rothstein picture of a farmer and his two sons during a dust storm in Cimarron County, 1936)



For my comparatively less-ambitious project, I created a fairly unrestrictive shooting script to guide my picture-taking in downtown Davis:

What are the representative businesses? The architectural styles? The landscaping? What transportation infrastructure is present? How are streets laid out and organized? How are open spaces defined? What sort of public art is there? Public gathering spots?

Following Jon Wagner’s 1979 photographic study of a suburban community in the U.S. (he also happens to be my professor for this class!), I also left open the possibility of capturing shots of whatever was visually-interesting to me. In this way, I allowed room for Bateson’s “artistically informed” strategy for data collection.

I began shooting at 7AM on a weekday morning and followed a serpentine but otherwise unstructured path through the grid of downtown Davis, crossing streets when I saw something interesting or wanted to frame a shot in a certain way. I limited myself to 2.5 hours, over which I took over 200 photos with a Nikon D90 (digital SLR).

Several themes emerged from this scripted photographic excursion. The photos included below are just a sample - they have not been cropped or altered in any way.

Businesses and business signs...

...both chains

and local spots.

Banks - lots of banks!

Emblematic town of Davis artifacts

and street signs,

many of which refer to bicycles


because there are a lot of bicycles in this town

I took lots of shots of bikes on porches as well as the ubiquitous Burley baby-carrying trailer
(the defining symbol of Davis, according to some).

Of course, I have some pictures of cars in Davis as well - there are plenty of them (and lots of Priuses),
but I didn't take nearly as many shots of cars as I did with the second method (you'll see).

Open spaces


Public places


Lots of public art


And some "unsanctioned" public art as well - all very small-scale and pretty tame, if you ask me.

And then there were some aspects of the town that may not have been representative but were
nonetheless interesting to me because I'd never really noticed them before.
For example, I found quite a few shut-down businesses

and old, abandoned telephones - artifacts of a different time, long ago

I also took seveal pictures of things inside store windows (again, you'll notice my tendency to photograph
food-related items - my personal bias of what is "interesting")

And here are a few other examples of images that, though aesthetically-pleasing, don’t necessarily
provide much information about Davis in and of themselves.

The photos taken with this method (shooting script + images of interest) might lead one to think that Davis is a super-colorful town full of restaurants and bicycles (and banks) and crazy, mosaic art and not much else. What happened to be visually interesting to me was very often something that I had never noticed before in all my years of bicycling quickly through town on my way to work or school or perhaps to grab a bite to eat. The things that I noticed, therefore, were not necessarily the same things that someone new to Davis might see or think of as representative.

Now let's see what Davis looks like when I tried to be a bit more scientific and objective about collecting photographic data.

Method 2: Walking with an algorithm and randomizing photographic decisions

You all know I’m no stranger to scientific research. I’ve struggled through statistics and understand the importance of randomized sampling procedures for reducing bias in all kinds of studies, from social surveys to water quality monitoring. However, I must admit that until this class I never dreamed anyone might actually apply randomized sampling protocols to visual research.

In “Avoiding Error” (1979), Jon Wagner (my professor again) describes the numerous ways he and his research team attempted to reduce bias and the influence of a particular photographer’s style while collecting photographic data about a suburban community in the U.S. I was particularly inspired by the detailed description of how they supplemented the more subjective images taken by photographers following “shooting scripts” with more randomized photographs. They randomized the picture-taking process through the use of a simple coordinate model and random number generators that dictated where each photo would be taken and its angle of orientation. While the random photos didn’t provide a significantly different view of the community from those taken in a more subjective way, they did direct the researchers’ attention to several features that they had previously neglected, recorded poorly, or completely overlooked.

This seemed like a fun thing to try; however, I did things a little differently.

A while back, I read about groups of people who call themselves psycho-geographers (as in psychological rather than insane although that can probably be debated, depending on the individual). These folks explore (mainly urban) places by walking around according to an algorithm (e.g. turn right, go two blocks, turn left, go one block, turn right). The practice is called dérive or “drift.” The idea is that most of us never deviate from the well-worn paths we travel in our home towns (or cities) on our way to work, our favorite restaurants, or the supermarket. As a result, we often miss many interesting, disturbing, or forgotten parts of the places we live.

While it’s doubtful that Davis has many deep, dark secrets, I thought that the idea of a walking algorithm would be a good way to randomize my path through the downtown grid. Here’s the algorithm I used:



Go straight. After two doors, cross the street, take a picture. Continue in the same direction as before, turn the first corner (or anywhere that leads through to another street). After two doors, cross the street, take a picture. Repeat.

If I came to the edge of the grid (B-G and 1-4 streets), I took the picture without crossing the street and then continued with the algorithm.

This algorithm allowed me to explore almost all parts of the grid between B-G and 1-4 and miraculously wound back on itself after about 2 hours of walking. I picked the starting point by placing a spinner on a map of downtown Davis - I started where the line of the pointer intersected the grid boundary on the map. A spinner, you ask?

Yes, instead of using a random number generator to randomize all of the other decisions involved in actually taking the picture, I made a spinner:


When I arrived at the spot designated by the algorithm to take a picture, I put the spinner in the middle of the sidewalk and spun it three times.

Spin 1: Determined the direction of the photo. I simply faced whichever direction the pointer did.
Spin 2: Determined the zoom between 18-105 mm.
Spin 3: Determined whether I pointed the camera angle up 20 degrees, kept it at eye level, or shot it while kneeling down.

I kept the camera on auto-focus and with automatic aperture and focal length settings.

Luckily Iago came with me and was my official “spinner.” This was a good thing because there was still a surprising amount of subjectivity involved in reading the spinner output and deciding what constituted a “door.” He also looked at the second-hand on his watch and told me when to take the picture (exactly on the next 15 second interval). When I complained that I was missing a really great shot because I had to point up or kneel down or turn in the opposite direction, he made sure that I stuck to the algorithm. (I admit I would have been tempted to "cheat" in some of these instances; having a witness kept me honest.) It was also a really fun time, even at 7AM in the morning!

I took 45 photos in about 2 hours. Here is a sampling of them (again, these haven’t been altered or cropped in any way).

Businesses again, although this time it's sometimes hard to tell exactly what they are...

Parking lots and street scenes were much more prominent in this group of pictures.

As were close-up pictures of walls...

...and shrubbery.

Cars also appeared more frequently, while bikes were almost completely absent.

One amazing thing happened on our random walk.
Remember this picture of the shoe repair shop and blue van from the first photo-shoot?

I kid you not, during the randomized walk, I took this almost identical picture:

It's a pretty small town afterall.

Perhaps the greatest difference in the two sets of photos was the number of tree shots. In the first set of photos, I took only one shot into a particularly beautiful canopy. During the randomized walk, half of my pictures were (almost) entirely trees.

If we lived in a city and the spinner had asked us to shoot upward at a 20 degree angle, we might have been likely to capture buildings or perhaps sky. Not in Davis. I realize the lack of trees in my first photo-shoot likely means that I take for granted the tall, shade-giving canopy that blankets the town.

A few final thoughts...

Let's revisit the questions I posed earlier.

Which method do you think more accurately represents downtown Davis?

Neither method on its own seems to provide a particularly accurate or complete picture of the town. Viewing both sets of photos together certainly does help, as would including more photographers in data collection. I also think a combination of the two methods could result in a fairly representative set of photographs. For example, I could use an algorithm to direct my walking and determine the locations of my picture-taking, coupled with a shooting script that would grant me freedom to take pictures of what is interesting at each stop.

Which method produces the most informative images? (i.e. If someone were presented these images 100 years from now without knowing anything about the town or society, which would give them a better idea of what life is like in this place?)

The first method (script + capturing visually interesting images) certainly produced photos with more information about the economic culture of downtown Davis - what people buy, eat, and sell. Those images also provide information about what people living in Davis in the early 2000's (such as myself) think is interesting to photograph. However, it really all depends upon the questions that you're hoping to answer with the images. The photographs from the first group left out the fact that while a lot of people bicycle in Davis, just as many (if not more) actually drive. I bicycle - bicycles are interesting to me, cars - not so much. The images in the second group remind us that there are still plenty of cars (and parking lots) in our bike-friendly town.

Which method produces the more evocative or visually-interesting images?

While I'd like to think that the images that I took with the first method were the most "visually arresting," I'm sure this is not always the case. After my presentation of this project to the class, Jon argued that the other images may be more "arresting" due to the fact that they are taken from unfamiliar perspectives. These images allow us to see Davis, quite literally, from a different angle.

What do you think?

References
Mead, Margret and Bateson, Gregory. (1977). On the use of the Camera in Anthropology. Studies in the Anthropology of Visual Communication. 4 (2): 78-80.

Schwartz, Dona. (1989). Visual Ethnography: Using Photography in Qualitative Research. Qualitative Sociology, 12 (2): 119-154.

Suchar, Charles S. (1997). Grounding Visual Sociology Research in Shooting Scripts. Qualitative Sociology, 20 (1): 33-55.

Wagner, Jon. (1979). Avoiding Error (pp.147-160). In Jon Wagner, Ed. Images of Information. Beverly Hills, CA: SAGE.