This site is archived. For my brand new live site go to Struggle Forever!

19 February 2010

Everything is Data: Latour's Four Notebooks


I'm still working my way through Latour's Reassembling the Social. Life's been pretty busy this past week, so I haven't had time to get through it - plus there's all the other reading I have to do for classes that's starting to pile up.
Anyway, going through the section on the "Fifth Source of Uncertainty: Writing Down Risky Accounts" I was intrigued by his idea of keeping four notebooks for research. To a degree it mirrors what many ethnographic methods books teach - you have your jottings notebook, your field notes, and your journal - but Latour has a slightly different idea of what the notebooks should include. His main idea is that "everything is data" and so everything about the research must be documented. Each notebook, therefore, is meant to capture a different aspect of the research process, and I want to describe them briefly here:

The First Notebook - The research log
In this notebook we would keep a detailed account of how we go about doing the research. We would note appointments, discussions with advisers, reactions to research from others, phone calls, internet searches, etc. - anything that describes how we came up with the idea for the research, how we get funding and from where, and how we go about finding the data. He says that "...[E]ven years after, it should remain possible to know how the study was conceived, which person was met, what source was accessed, and so on, at a precise date and time."
This is reminiscent of a lab notebook where you are supposed to describe in detail every step you took in performing an experiment. They serve two main functions: 1) if something goes wrong, you can look back and see where things might have gotten mixed up and 2) others can look at your work and replicate it exactly. In fieldwork, we generally can't replicate our research, but by documenting every step, we can show how we got our results or look back and see where we might have gone wrong.

The Second Notebook - The Data
This notebook is where we keep our data - interview transcripts, field notes, thoughts, observations, etc. Latour says that it should be "possible to simultaneously keep all of the items in a chronological order and to dispatch them into categories which will evolve later into more and more refined files and sub-files." This is really easy to do with computers, and I think Qualitative Data Analysis (QDA) software like Atlas.ti or TAMSanalyzer is the best option. QDA software allows us to tag our text at the paragraph or even single word level (as opposed to tagging the whole text in EverNote or other bookmarking type programs). That way we can either scroll through our data as a whole chronologically, or we can query those sections that we've tagged with particular codes. We can even call up sections that are cross-referenced with two or more tags for more intricate analysis.

The Third Notebook - Writing
Latour says that we should not wait until the data is collected and the research is finished to begin writing up our results. This is what the third notebook is for - writing up the research as we go. If some idea comes to us as we're collecting data, some speculation on what might be going on, we would write it here. Think it through, play with it, experiment with writing, with using different models and metaphors. Not only is this good practice for the sake of the research - to keep your own thoughts from mixing with those of your informants - but it will also improve your writing.
For this notebook, it occurred to me that the best resource might be a blog. I say this for two reasons - it's a good place to write casually about anything we like and keep track of our work, and it makes our research available to a large audience. Other researchers, the general public, or our informants themselves can come by and see exactly what we're thinking. They can even comment, add their own thoughts, critique our thoughts or challenge us to dig deeper. I know the comments on this blog often help me think more about whatever it is I'm writing on, and at times they have changed my whole way of understanding a topic.

The Fourth Notebook - The Effect of the Research
In my previous post, Mediators and Intermediaries, I mentioned that Latour sees every part of the research as having an agency and value of its own. We can no longer think of the research divided into discrete phases - preparation, data collection, analysis and write-up. Instead we must see them all as interwoven with one another. Furthermore, the whole process doesn't stop with the write-up - it continues as the effects of the research and the report (small though they may be) cascade outwards to affect others. The fourth notebook, according to Latour, should document the effects of the research. Even if the report is only read by one or two people, that should be noted and its trace should be followed. This is not an exercise in vanity - it allows us to understand what happened with our research after the project was finished, it allows us to see what people are doing with it, how it has changed behavior (or not), and how we could do a better job of making our research useful and effective.
Last semester, when I was taking my applied anthropology course, we read article after article in which the researchers said that they had created a website for some (noble) reason. Time after time, we would visit the website only to find broken links, dated material and a general sense of waste. Too often these grand projects are completed and then forgotten - allowed to decay - without a second thought. Keeping track of what happens to our research after it's done is essential to keeping it relevant and worthwhile.

Those are Latour's four notebooks, and, like I said, I find it absolutely wonderful. I think if we all followed his advice, it would make us better researchers, better writers and better activists. I think it applies equally well to those of you who aren't social scientists, but perhaps want to write literature or make films. Maybe you don't have to keep all four journals, but keep in mind the axiom "Everything is data" and do what you can to document any and every aspect of your life including interesting stories, mundane details, unique ideas, and observations, . You may not use all of it, or even any of it, but I assure you that it will have a positive effect on your skill as a writer and on your general creativity.
That said, it's quite a tall order. For someone like me who has to take classes, teach classes, run from place to place, cook food, write papers, read articles and books, not to mention find time for some minuscule social life, I think I'd find it difficult, if not impossible, to maintain all of these notebooks for all of the projects I might work on. It's a worthwhile goal, though, and I will certainly do my best.

8 comments:

idyllhands said...

I've always had trouble keeping my notebooks organized. This year, I've been trying to keep one central notebook where all of my rough drafts go, and have placed several notebooks around the house for use in jotting down ideas, or what I call "sketches" notebooks. This is so that I don't end up with 5-6 notebooks floating around for a project I'm working on, I should just have 2: one with the sketch in it, then one with the rough draft. Of course, I haven't been very productive yet this year; I've been too busy with work and buying a house and all.
I'm not sure who said it, but I liked the quote: that a good system for organization is any one that you actually use.

idyllhands said...

Just wanted to add, going on your everything is data premise, that I seem to remember reading that Ray Bradbury kept a notebook of ideas since he was young, and would often (sometimes years later) write stories based on those jottings.

btmc said...

My notebook methods are very sporadic. I completely agree that writing whatever is on the mind as often as possible only makes everything written better. I almost never practice this, but when I have it is sooo good.

I think the main thing I find interesting about Latour's method is the separation and categorization of the elements of research. I have a really hard time separating things into categories long enough to work in that frame of mind. But I imagine that other minds work very well with order and definition.
I would ask, what is the benefit of separation, and is it something I need to learn. Is it simply to get us to examine the work from these four angles? Could we then say that even more notebooks would only increase the efficacy of our work? When does the method become more complicated than the subject demands?
Essentially my feeling is that this seems to be a blueprint for doing good work. But I think it must be customized, somewhat, to anyone who wants to use it.
Good post.

Jeremy Trombley said...

Josh, I agree completely with your last statement - that the best system is the one you actually use. No doubt any of these ideal systems will be adjusted and adapted to fit one's own needs and circumstances.

Brendan, I would say that the separation is useful methodologically for the social sciences. It's true that everything is data, as he says, and that's why everything must be written down. But everything is not the same kind of data. Mixing your interview data with your methodological notes and your writing up would only make analysis more complicated at the end. There are also issues of confidentiality to consider if you're using a public space such as a blog to keep track of things - you want to keep the names private, though you might make the data accessible.
The separation is also useful as an illustration since it allows us to see the different parts of the data and how much we should actually be recording. That, in itself, doesn't require physically separate notebooks, but for the sake of talking, it's easier to talk of four notebooks.
For most people, I think one or two journals would be fine. Again, what Josh mentioned holds - the best system is the one you use.

I'm working really hard right now to write a lot more. I've set a non-binding goal to write one EI post per week and one Prism post per month. I'm also keeping a daily journal again, plus there are school papers and whatnot. I think in the short time I've been doing that, my writing has improved - my last few EI posts, I feel, are some of the best I've done. The challenge is to maintain it over a long period of time. It's a worthwhile goal though.

J.M said...

If it can encourage you, let me add that I agree : these last two posts are particularly well written, clear and focused (no flattery here).

In some books I've read, researchers made a distinction between a "field notebook" (which is the equivalent of Latour's second book), and a "research notebook" (which is the equivalent of Latour's first and maybe third notebook altogether).


Although I agree that having different kind of notebook for different kind of writing, I must admit I am a bit confused about this kind of formalisation. To take an example, should the description of the context of an interview (be it formal or not) be included in the "data" notebook, or in the "research log" ? What would it change ?
Another example, if you suddenly come up with a good way to describe what happens, will you write it in the "writing notebook" and then go for your research log and let a note about this new development of your research ? (It could be a good way to control one's research, but I guess it would demand a very strict discipline that is not that essay to comply with).

What I find really new is Latour's idea of the fourth notebook, and I realize that until now, this one exists only vaguely in my head, and that I should write things down.

And I certainly agree with Latour that one should not wait until the "end" to start "writing up the results".

Jeremy Trombley said...

Thanks J.M. I'm glad you like them too!
I think there's definitely some ambiguity in the definition of the notebooks, and I definitely agree that it would require more discipline than I could probably muster. It's a worthwhile goal though - an ideal to bear in mind, perhaps.

I'd say that it's not so much about a different kind of writing, but a different kind of data. In the first notebook, the data should answer the question "How did I go about doing the research?" That way you and others can look back and see what you got right or wrong. It's a methodological notebook. The data in the second should answer the question "What is my data?" or something like that. The context of an interview could go in both depending on whether or not you feel it's relevant. The third notebook should be your findings, but rather than simply writing it up as an article or thesis at the end of the research, Latour would have us write as we go. It should answer the question "What have I learned?" The fourth notebook is the most interesting to me as well - as you say Latour's novel contribution. It should answer "What has become of my research since it was published?"

As I said, there's ambiguity, and flexibility, I think. There's also what Josh mentioned: the best system is the one you use. I'm glad he mentioned that, cause I think it's very true.

BTW - I'm working on a Prism post too; I should have it up by the end of the week.

J.M said...

Thanks for your answer.
Yes, there is ambiguity, certainly due in part to the fact that we can consider that "everything is data", but that at the same time, we have a notebook (among 4) dedicated especially to "data".

But let's just say there is different "levels" of data.
And yes, let's use the trick we're the most comfortable with.

I'll definitely read your Prism post when you're done.

Jeremy Trombley said...

To be fair, the "Data" label on the second notebook is my own, not Latour's. Here's what he says about it:

"A second notebook should be kept for gathering information in such a way that it is possible simultaneously to keep all the items in a chronological order and to dispatch them into categories which will evolve later into more and more refined files and subfiles" (p134 RTS, italics his).

To me that refers to what we normally call "data" - interview transcripts, field notes, survey results, etc. - and QDA software seems like the best system for organizing it according to his guidelines.

Related Posts with Thumbnails