It is currently held, not without certain uneasiness, that 90% of human DNA is ‘junk.’ The renowned Cambridge molecular biologist, Sydney Brenner, makes a helpful distinction between ‘junk’ and ‘garbage.’ Garbage is something used up and worthless which you throw away; junk is something you store for some unspecified future use. (Rabinow, 1992, 7-8)
Junk as Failure
In the bioscience lab near Tokyo where I did my ethnographic study, the researchers taught me how to do PCR experiments. This was before Covid when almost everyone came to know what PCR was, or at least, what kind of instrumental information it could be good for. The lab was working with mouse models, although I never got to see them in their cages. But the researcher I was shadowing showed me how to put the mouse tail clippings she collected into small tubes. She hated cutting tails, by the way, and preferred to take ear punches when she could. She told me that she didn’t like the way the mice wiggled under her hand, as if they just knew. But at this point anyway, the mice are alive in the animal room and she is only putting small, but vital, pieces of them into a tube to dissolve them down (mice becoming means), to get to the foundation of what she really wants.
I’ve still got the protocol that I typed up from the notes I made with her in the lab. Step 1 was: “Add 75 ul of NaOH to each ear punch tube (changing tips as I go).” The changing pipette tips part was really important to avoid haphazardly spreading around DNA, I learned. I also had to make sure the clippings were at the bottom of the tube and submerged. She said I could flick the tubes with my finger to get the “material” to fall down to the bottom and she showed me how to do it. I also, she cautioned, always had to be very careful of bubbles, but more flicking could help there and by making sure I didn’t put the pipette too far down into the solution. Then we would spin the tubes in the vortex (which I always typed as VORTEX for some reason), add some other reagents, and put it all in the “PCR machine,” but that is not at all its technical name. Then we would usually go with all the others to the cafeteria for lunch.
In writing this now, I couldn’t remember what “NaOH” stood for so I had to ask the internet. And as I looked back over this protocol, and these practices I was just barely learning to embody before the pandemic sent us all home, I realize that they must have settled back in my mind somewhere, just as the material-ness of the lab which anchored them for me has receded like a shrinking lake in a drought summer. But what I do hold on to is what the researcher taught me about the importance of repetition and focus, for a kind of purity of practice, and the diligence to make materials—whether of mice or of sodium hydroxide—do what they ought to do.
Because what captivated me about these initial PCR steps was what appeared to me to be the profound transformation they wrought (of course, I am not the first person to say so)—from fleshy ear punch to silt DNA multiplied in a clear plastic tube, with just a little bit of chemicals and some repetitive cycles of heat—but even more, how this transmutation had the potential to fail in one way, or for one reason, or another. How difficult it could actually be to get the materials, and even the researchers themselves, to do what they ought. Once, I used some unknown solution instead of water because it was on a shelf in an unmarked bottle close to where the water, which I later supposed had gone missing, was usually kept. Once, I didn’t remember to change pipette tips. Or the sense in my hands of precisely what to do next and properly would simply begin to unravel. When we had to throw the tubes in the trash, the researcher comforted me by telling me about a time when her mind wandered for just an instant while pipetting and she lost track of which tube she had last filled with reagent. A minor momentary mistake that grows, and can even burst, into a huge error in the downstream. She taught me that sometimes, if I lifted the tubes to the light to examine their volume of liquid, I might be able to get back on track. Other times the PCR machine might not cycle its heat properly. One machine was already considered to be of questionable working order but the lab didn’t have the funding to replace it. We didn’t know about its full potential for failure until we got all the way through to the very last stage of the process and discovered we had to go back to the beginning with new clippings.
Junk as Potential
The researcher and I classified these particular (wait, was that water?) experiments-in-the-making as failures because they missed the mark of their intentions. Their purposefulness, decided in advance by the goal of genotyping these mice, was also appended to other purposes, specifically to cultivate a living gene population that the researchers needed for other more central concerns. Trashing the experiments that deviated from this intentionality, although it could be costly, was a seemingly simple decision. After the PCR melt and the second half of the experiment, the electrophoresis machine either “read” back the base pair numbers we were looking for, or those numbers were just wrong and we’d made an obvious mistake. Or worse, everything collapsed into inconclusiveness and we needed to repeat the experiment anyway. In this case, deviation from expectation, and therefore from usefulness, was what pushed experiments to a kind of failure, beyond which point they could not, in this context at least, be so easily reclaimed.
But what does something like “junk” have to do with mice ear punches, chemical transmutation, and mundane laboratory failures? Garbage experiments are routine in scientific practice after all. But as any scientist might tell you, failure can be its own kind of productive; in the least, as a way to learn the value of steady hands, and how to recognize water by smell, or its necessity as a control in genotyping—to become a “capable doer,” as one scientist told me. But beyond these mundane errors, some scientists argue that failures of a particular kind can break open old ways of thinking and doing, although what that failure is, and can be, is variously classified:
Science fails. This is especially true when tackling new problems. Science is not infallible. Research activity is a desire to go outside of existing worldviews, to destroy known concepts, and to create new concepts in uncharted territories. (Iwata, 2020)
I wish “failure” were the trick to seeing and moving beyond the limits of current knowledge. Is that what Kuhn said? I think that paradigm change requires making a reproducible observation that does not fit within the existing model, then going back to the whiteboard. But I don’t think these observations are very well classified as a failure. If failure = unexpected result of a successful experiment/measurement, then I can agree. (Personal communication with laboratory supervisor, 2020)
Failure has more potential than we might often recognize, where an instinct to trash can instead push to new beginnings. Just as Rabinow described Brenner’s description (1992), failure is like junk, those materials or states that are in-the-waiting—waiting to be actualized, reordered, and reclaimed as meaningful, valid and valuable, even if we don’t yet know how or why. Junk is, in this way, more than matter “out of place,” although it may land there interstitially. If “[d]irt is the by-product of a systematic ordering and classification of matter, in so far as ordering involves rejecting inappropriate elements” (Mary Douglas, 1966, 36), then junk is garbage and failure and decay, and even breakdown, on the precipice of being made anew. After all, without intentionality or purposefulness and other values, there can be no garbage, or failed and failing experiments and paradigms, in the first place.
Consider an example that seems categorically different from scientific experiments: inventory management in role-playing videogames. In Diablo 4 (2023), any item picked up from downed enemies or collected in the environment can be marked as “junk” and then salvaged by visiting an in-game merchant. These bits of amour and other gear reappear in your inventory afterwards as junk’s constitute materials, useful again for crafting and building up new things—strips of leather and other scraps as well as blueprints for better stuff. In Fallout 4 (2015), the “Junk Jet” gun lets you repurpose your inventory instead as ammo, anything from wrenches to teddy bears, which can be shot back out into the world and at random adversaries, where you might later be able to pick them up again, if you want. Managing encumbrance in Skyrim (2011), on the other hand, is a task of drudgery and tedium. Almost every item in the game world is moveable, each with its own weight calculation, and can be picked up and stored even accidentally, until your character is weighed down to the point of being unmovable. But the game is designed to make you feel that there is always the possibility that some magical potion, random apple, or 12 candlesticks, might just come in handy for a future encounter, a book that you might really read later, leading to a hesitancy to trash anything. In turn, every item brims with, as yet undiscovered, use-value. As Caitlin DeSlivey argues: “Objects generate social effects not just in their preservation and persistence, but in their destruction and disposal” (2006, 324). And certainly this is true when, over-encumbered deep inside a dungeon, I agonize over which items to drop, in order to move again, in order to continue to collect more—or laugh as I spray the world with cigarettes and telephones.
For me then, junk is a way to look for when and where particular boundaries of the useful or valuable—and even the clean and functioning—are “breached” (Helmreich 2015, 187), and then reordered. Although Helmreich is speaking to scientific experimental practices and their organizing ideologies, his insight is useful for junk’s attention to those very breaches: “moments when abstractions and formalisms break, forcing reimaginations of the phenomena they would apprehend” (185). Of course, junk DNA itself has experienced this very kind of breaching—more recent scientific research demonstrating its non-coding role is actually not without usefulness (c.f. Goodier 2016)—(re)animating it for future use. And although DeSilvey is describing vibrant multispecies-animated decay within abandoned homesteads, like Helmreich, she points to junk’s transformative potential. We just have to dig through rotted wood and insect-eaten paper, or virtual backpacks and books, to find it.
Junk as Repair
Junk merges failure, trash, and decay with the processual and everyday negotiation of culturally meaningful and policed categories: garbage, scraps and waste, but also “breakdown, dissolution, and change” (Jackson 2014, 225). Although Steven J. Jackson describes the ways these last three are fundamental features of modern media and technology, an anthropology of junk collects and extends these processes into broader techniques and social practices. Junk can help us see connections criss-crossing symbolic and material breakage and disintegration. It helps us see in/visibility of the dirty and diseased, not as a property of any material or technological object alone, but as also always in coordination and collaboration with the ways they are imagined and invested—and more, always enmeshed in variously articulated forms of power.
If infrastructures like computer networks, for example, become (more) visible when broken (Star 1999), it is not their brokenness or decay in an absolute sense that reveals them, but the way their state change defies our everyday and embodied expectations—the way they push against normativity. We may be just as surprised to find things in good working order.
Bit rot after all, has just as much to do with the made-intentionally-inoperable systems that force the decay, or really uselessness, of data (Hayes 1998), as it does with any actual mold on CD-ROMS and other corruptions of age and wear. In fact, digital information or technological and material infrastructures don’t become broken, just as they don’t become fully ever fixed either. Breaks and breaches are hardly so linear. Instead, these are “relative, continually shifting states” (Larkin 2008: 236). This view may be in contrast to Pink et al.’s suggestion to attend “to the mundane work that precedes data breakages or follows them” (2018, 3), but not to their entreaty to follow those everyday practices of maintenance and repair, and even intentional failure and forced rot. This is not simply because data and other material practices like PCR experiments may fail under given conditions or focused intentions, perhaps as a result of a momentary distraction or a faulty machine—or in the case of programming, because debugging is actually 90% of the work, as one bioinformatician told me. Indeed, software testing in practice goes beyond merely verifying functionality or fixing bugs and broken bits of code, but helps to define and make “lively” (Lupton 2016) what that software is, and can do, and can be made to do in the first place (Carlson et al. 2023). Along the way, as a generative process, testing, tinkering, and fixing have social effects (DeSilvey 2006) which are external to, but always in extension of, broken/working materials themselves (Marres and Stark 2020).
Junk as Resistance
More importantly, perhaps, broken things can be used, as Brian Larkin argued in relation to Nigerian media and infrastructures, as a “conduit” to mount critiques of the social order (2008, 239)—to call attention to inconsistency and inequality, and to demand or remodulate for change. To see this resistance at work demands a collating of junk practices. As Juris Milestone wrote in his description of a 2014 American Anthropology Association panel, “What will an anthropology of maintenance and repair look like?”:
Fixing things can be both innovation and a response to the ravages of globalization—either through reuse as a counter-narrative to disposability, or resistance to the fetish of the new, or as a search for connection to a material mechanical world that is increasingly automated and remote.
Junk’s transformative potential asks us to see removal and erasure, or in Douglas’ terms “rejection,” as always coupled to these reciprocal practices: rebirth, repair, repurpose, renewal. In this way, junk shows us the way decay, even technological corruption, is less a “death” than a “continued animation of other processes” (2006, 324).
But if junk describes a socio-cultural ordering system concerned with practices of moving materials—even ideas and people—into and out of categories of value and purposefulness, it must also contend with the vital agency of other material and microscopic worlds, which just as easily unravel out or spool up regardless of human presence, intention, and desire. Laboratory mice in fact are particularly disobedient, they hardly ever behave as they are supposed to—just as cell cultures in a lab are finicky and fail to grow to expectations, and junk ammo from the Junk Jet has a 10% chance of becoming suspended in mid-air, becoming irretrievable. If we repurpose sites or moments of breakdown to resist configurations of power, then materials themselves are also always resisting what they ought to do or become. This is the draw of the things in which we are enmeshed, where we are always extending, observing, destroying and deleting. If junk is the possibility, under particular cultural expectations and desires, for things to be pushed or cycled across such thresholds, and also, of making and unmaking these, it also must contend with the things themselves—with what we see in a corroded mirror, looking, or not, back at us.
Although junk may be over-bursting in its use here as a metaphor, I argue it can still usefully be used to stitch growing anthropological attention to material decay, breakage, and deviation together with tinkering, maintenance, and repair—across locations, states, practices and materialities. Granted, “manifesto” is also a too decisive word to attach to this short piece. Too sure of itself. But this post is also an attempt to challenge the understanding of what it means to be (academically) polished and complete. I use manifesto here mostly tongue-in-cheek, while still holding to the idea that any argument has to begin in small seeds, and start growing from somewhere.
My thinking about junk began years ago with Brian Larkin’s attention to breakdown (2008). More recently, I found DeSilvey (2006) by way of Pink et al. (2018); and Jackson (2014) from Sachs (2020); and Hayes (1998) from Seaver (2023). This lineage is important because I am not inventing, but building. These ideas are also bits and tears of conversations with Libuše Hannah Vepřek, Sarah Thanner and Emil Rieger, and very long ago, Juris Milestone. But everything gets filtered first through Jonathan Corliss.
This research has been supported by the Japan Society for the Promotion of Science’s Grant-in-Aid for Scientific Research (C) 20K01188.
 PCR stands for polymerase chain reaction. It is an experimental method for duplicating selected genetic material in order to make it easier to detect in secondary experiments.
 Thermal cycler, for anyone interested. Also, just to note, but for the purposes of this retelling, I gloss over the most detailed part in writing so simply: “add some other reagents” and later, “after the PCR ‘melt’ and the second half of the experiment.”
 I wrote in my protocol notes, as an (anthropological) aside to myself: “K. stressed that the amount of liquid in this case doesn’t have to be super accurate, but that this is rare in science experiments. When I tried it for the first time, I almost knocked over all the new tips and also the NaOH solution which can cause burns! Yikes~)”
 Inconclusiveness includes an unclear or unaccounted for band in the electrophoresis gel, which is seen in the machine’s output as an image file.
 The images in this post are part of the artistic work of Sarah Thanner, a multimedia artist and anthropologist who playfully and experimentally engages with trashing and untrashing in her work.
 Fallout Wiki, Junk Jet (Fallout 4), https://fallout.fandom.com/wiki/Junk_Jet_(Fallout_4)
 Here, I also gloss over (new) materiality studies, Actor Network Theory, etc. which have linages too long to get to properly in this small piece.
Carlson, Rebecca, Gupper, Tamara, Klein, Anja, Ojala, Mace, Thanner, Sarah and Libuše Hannah Vepřek. 2023. “Testing to Circulate: Addressing the Epistemic Gaps of Software Testing.” STS-hub.de 2023: Circulations, Aachen Germany, March 2023.
DeSilvey, Caitlin. 2006. “Observed Decay: Telling Stories with Mutable Things.” Journal of Material Culture 11: 318-338.
Douglas, Mary. 1966. Purity and Danger: An Analysis of Concepts of Pollution and Taboo. London: Routledge.
Goodier, John L. “Restricting Retrotransposons: A Review.” Mobile DNA 7, 16. https://doi.org/10.1186/s13100-016-0070-z
Hayes, Brain. 1998. “Bit Rot.” American Scientist 86(5): 410–415. http://dx.doi.org/10.1511/1998.5.410.
Helmreich, Stefan. 2015. Sounding the Limits of Life: Essays in the Anthropology of Biology and Beyond. Princeton: Princeton University Press.
Iwata, Kentaro. 2020. “Infectious Diseases Do Not Exist.”「感染症は実在しない」あとがき. Retrived May 9, 2020, https://georgebest1969.typepad.jp/blog/2020/03/感染症は実在しないあとがき.html.
Jackson, Steven. J. 2014. “Rethinking Repair.” In T. Gillespie, P. J. Boczkowski, & K. A. Foot (Eds.), Media Technologies: Essays on Communication, Materiality, and Society. Cambridge: MIT Press. Pp. 221-239.
Lupton, D. 2016. The Quantified Self: A Sociology of Self Tracking. Cambridge: Polity Press.
Marres, N, Stark, D. 2020 “Put to the Test: For a New Sociology of Testing.” British Journal of Sociology 71: 423–443. https://doi.org/10.1111/1468-4446.12746.
Milestone, Juris. 2014. “What Will an Anthropology of Maintenance and Repair Look Like?” American Anthropological Association Meeting.
Pink, Sarah, Ruckenstein, Minna, Willim, Robert and Melisa Duque. 2018. “Broken Data: Conceptualising Data in an Emerging World.” Big Data & Society January–June: 1–13. https:// doi:10.1177/2053951717753228.
Rabinow, Paul. 1992. “Studies in the Anthropology of Reason.” Anthropology Today 8(5): 7-8.
Sachs, S. E. 2020. “The Algorithm at Work? Explanation and Repair in the Enactment of Similarity in Art Data.” Information, Communication & Society 23(11): 1689-1705. https://doi:10.1080/1369118X.2019.1612933.
Seaver, Nick. 2022. Computing Taste: Algorithms and the Makers of Music Recommendation. Chicago: University of Chicago Press.
Star, Susan Leigh. 1999. “The Ethnography of Infrastructure.” American Behavioral Scientist 43(3): 377–391. https://doi:10.1177/ 00027649921955326.