Policy, Research, Outreach: Environment, Food & Biosecurity
Home » Research » Actor-Network Theory » How ANT Led to Unexpected Results
How ANT Led to Unexpected Results

When regulatory and artisan practices and the objects that each involve are observed at close range—as described on the previous page—the differences between them become more difficult to pin down than one might expect. In fact, it is perilous to juxtapose the notions of craft production and regulation, or even the notions of craft and industrial production (see also Lowe and Ward 1997). Artisans make use of objects that their supporters may dismiss as industrial, and some of them follow—at their own initiative—regimented procedures suggestive of operations engineering. And food safety inspectors—the putative foot soldiers of industrial food—carry out their work in a manner that may be lauded as artisan. This section explores these two observations.

The Craft of Inspection

Food safety inspectors ply a craft. True, the practice of inspection draws upon a large body of scientific knowledge and an accumulation of statutes, regulations, and other policy documents. Yet conducting an inspection calls upon expertise that cannot quite be put into words—that is, tacit knowledge—quite apart from what is conveyed in codified law or agency procedures (Polanyi 1958). Inspections are adaptive and interactive processes that are contingent on the facility being inspected, the people with whom the inspector engages, and the inspector’s own skill. Inspectors work independently, and their tools are both few and technologically simple.

In fact, the “task orientation” of pre-industrial work routines described by historian E. P. Thompson (1967)—much more than the productive and bureaucratic efficiency of industrial work—characterizes the practice of inspection. It does so in several respects. First, many of the inspectors in this study took whatever time was required to communicate with producers and to help them understand the justification behind regulations. Second, they did not clearly distinguish between “work” and “life” in their interactions with artisans. They chatted about children, pets, and hobbies as well as about regulatory compliance. They “made themselves human” and approached artisans on a human level as well, looking “past the business,” in the words of one inspector. Third, this approach to inspections may appear to an outside observer to be an unreasonable and inefficient use of time. The time that inspectors spend in facilities cannot be evaluated in standardized (i.e., industrial) terms—not in inspector hours per square foot of facility, per unit of production output, or even (if it were possible) per microbe. Measuring efficiency would be futile and at cross-purposes with the “logic of need” according to which many inspectors operate (Thompson 1967, p. 59).

This section explores four of the ways in which inspectors’ work mirrors that of artisans: in embodied skill, variable and adaptive work, autonomy, and technological simplicity.

Developing Embodied Skill

Inspectors learn their jobs through professional and life experience. The practice of food safety inspections requires knowledge and skills that are not conveyed in regulations themselves or through an understanding of the scientific basis for regulatory requirements. Translating regulations so that producers understand and respond to them involves communication skills, discernment, and patience.

Artisans and inspectors both spoke of the difference that age and experience make in an inspector. One middle-aged inspector stated firmly that if he had started directly out of college, he “absolutely would have failed at this job” because he saw the world in black and white. Younger inspectors were generally characterized as stricter, more “by the book,” and harder to work with. “Being yourself” and establishing interpersonal connections were important. Inspectors who had children or who had taught school spoke of the difference those experiences made in the greater flexibility and understanding that they brought to inspections: “If you don’t take the time to turn yourself into a human being [for producers], then they see you as a bad parent.”

Variable and Adaptive Work

Inspection is adaptive work that is difficult to standardize. There is tacit knowledge (Polanyi 1958) even in what appear to be regimented activities. Asked what made someone a good inspector, a dairy inspector began his response stating, “Flexibility as things come up that you have to react to.”

Inspectors “make” policy within the varying situational contexts that their field work presents. In the many instances in which a violation is not classified “priority,”[1] inspectors have broad discretion in deciding whether to record it. “You don’t want to overload people,” one explained, “I never write up everything that I could.” Their interaction with producers is integral to the practice of inspection:

If there’s no one there, you have to write down everything you see. [But if] the producer is there and you have a conversation, there are things that you don’t need to write down.

Dairy inspector

Many of the inspectors in this study spent considerable time talking with producers in order to learn about their operations and their approaches to production. They indicated that artisans, like many small plant owners, rely on them as quality control consultants; smaller food facilities may not have the resources to hire specialized staff. Inspections involved lengthy conversations about issues only tangentially related to food safety. “Walk in their shoes,” one of them described her approach to her job. These discussions with producers built trust, established inspectors’ credibility, and enabled inspectors to better exercise their discretionary decision making:

[The interpersonal rapport between the producer and the inspector] allows the humanness to come through in a program that’s designed to remove it. All of the regulations…are designed to remove the decision making process, remove interpretation, remove humanity, and make it a cut and dried, black and white, “do not think.” Nobody likes life like that. On either side. Building a relationship allows both parties to address the absurdity of it while seeking a mutually satisfactory experience.

Jam producer

Regulatory outcomes are affected by what the inspector knows of a producer’s situation, by a producer’s receptiveness to inspector directives, and by the trust and collegiality that they have established. During one bakery inspection observed in this study, the inspector’s decision whether to record non-priority violations was based on her knowledge of the bakers’ situation and her past experience with them. She did not record the violations that she was confident would be corrected, and she recorded violations that she felt might require extra incentive.

Outcomes are also affected by inspectors’ experiences at other facilities. Inspectors indicated that their degree of familiarity with artisan facilities affected their interpretations of artisan situations. In some cases, as artisans explained their methods and the equipment they used, inspectors adjusted their understanding of concomitant health risks and, hence, their assessment of what constituted a violation. Some food inspectors who had initially been reluctant to accept the use of linen-lined wicker proofing baskets in bread baking came to view them as an acceptable part of artisan processes.


Inspectors work independently, both physically and as decision-makers. They work out of state vehicles and manage their own time. Vehicles serve as mobile offices, containing computers, printers, and paper files. Many meals are eaten on the road. True, inspectors’ decision making is constrained. (And the same may be said of medieval artisans in guilds.) And the practice of inspection is founded upon a large body of scientific knowledge and an extensive set of statutes, regulations, and other policy documents. Still, inspectors internalize this information; they make decisions independently in enforcement situations. When an inspector is not sure how to address a situation, he or she calls on other inspectors or supervisors for advice. Regional supervisors convene periodic meetings of field staff, and field staff meet annually for a statewide training.

Technological Simplicity

The implements that inspectors use in conducting food safety inspection are few and technologically simple. During the inspections observed in this study, implements included clipboards, paper, and pens; flashlights; thermometers; and paper strips for testing sanitizer solutions. Inspectors took notes by hand, often in pocket notebooks, and sometimes checked boxes on paper forms. Following inspections, they entered their findings into evaluation reports on laptop computers either in the facilities or at another location. These reports form the official and public record of a facility’s compliance status and provide a reference for future evaluations.

[1] Priority violations represent immediate health risks to consumers. Examples include insect infestations and refrigerator temperatures that are too high.

The Industrial Technology of Artisanship

When observed at close range, the practice of artisanship has elements of industrial production. Examining the role that these industrial elements play sharpens our picture of artisanship in a contemporary, industrialized, regulated context.

Artisan food processors in the contemporary US make use of manufacturing technologies and standardize some aspects of their production operations (e.g., Paxson 2012). Even where not required by regulations, most of the artisans in this study used some form of motor-powered technology, and many of them kept records beyond what regulations required. In other words—to the extent that artisanship is defined by a reliance on manual, pre-industrial methods and an intuitive embrace of variability—much contemporary artisan practice is not “artisan.” This is not to say that artisans fail at their own craft. Instead, broad characterizations of artisanship as “non-industrial” fail to do justice to the skill, adaptability, and autonomy that artisans exercise within a contemporary industrial context. This section examines artisan use of industrial technologies to preserve food and reduce labor, and it considers the place of standardization and industrial technologies within a logic of craft production.

Industrial Technologies

Artisans use manufacturing technologies in several ways, including to control temperature during production processes, to limit microbial growth, and to reduce labor.

Some of the bakeries visited in this research used convection ovens rather than wood-fired brick ovens, although the law allows brick ovens. Convection ovens are heated electrically or with natural gas and require less time and labor than do brick ovens. Those who used convection ovens produced bread in quantities that made brick ovens impractical.

All facilities visited in this study used mechanical refrigeration—refrigerators and freezers. Refrigeration controlled fermentation for bakers and cheesemakers. Some bakers used refrigerated space as dough retarders, in which the temperature slowed the development of bread dough until bakers were ready to allow it to rise. Cheesemakers managed the temperatures of aging rooms, or caves, by mechanical means. Refrigerators and freezers prevented unwanted microbial growth in ingredients and finished products.

Mechanical mixers and stirrers were commonly used as labor saving devices. All bakers used automated dough mixers, and nearly all cheesemakers used milk pumps or automated stirrers. This enabled artisans to produce larger quantities of product and freed up time for other tasks (see Paxson [2012] for similar findings). In some instances, this machinery prevented strain on employees’ bodies. Taking dough out of mixers involves bending over and extracting large volumes of dough by hand. One baker had installed a tipping mechanism to reduce bodily strain. He next hoped to install a machine to raise and lower the facility’s oven loader in order to prevent employee shoulder strain.

Cheesemakers discussed the relative merits of stirring curd by hand or with a mechanical stirrer. Nearly all of the cheesemakers in this study used mechanical stirrers. Describing the interaction of machinery with the curd, this cheesemaker deemed machinery to be acceptable at certain points in cheesemaking if it would not “do any more damage:”

When you take milk in cheesemaking, you’re damaging it the whole time. Everything you do is going to do some damage…The only reason I hand stir it is, it’s on such a small scale that a mechanized arm doesn’t make economic sense. If I had that mechanized arm, I would still do the initial first 20 minutes’ stirring by hand, because the curd is very soft and fragile and the arms might damage it. After that, it’s firmed up, and the arms won’t damage it.



Contemporary artisans manage a delicate balance. On one hand is the variability of working within ambient temperatures and humidities and using ingredients that may fluctuate throughout the year. On the other hand, artisans set quality standards and work to attain them, and they produce for markets that may frown on surprises. Time and temperature controls and careful recordkeeping helped meet and monitor artisans’ standards.

In fact, the regimentation and recordkeeping of some artisan processes suggested the rigors of operations engineering. Artisans felt that the creative dimension of craft production has been overstated in popular accounts. “[You] don’t survive it, if you don’t have that attention-to-detail side of you,” stated a farmstead cheesemaker.

Industrial Technology and Craft Logic

Artisans use industrial technology, but they do not adopt an industrial logic of production. The difference between industrial and craft production lies in part in the meanings that artisans ascribe to the technologies that they use.

Artisans put industrial technology to the service of craft production. They use machinery selectively. For example, the tipping mechanism mentioned above served workers by reducing bodily strain. In another bakery:

I use a mixer that does all the heavy lifting for me. And that’s certainly not handcrafting. But there is a style of mixing that I employ that is not cookie cutter mixing. I mix to touch, to feel, to temperature, to ration…If I undermix, then I know that at the bench I have a lot more additional work to do over the next few hours or the next 24 hours. That’s certainly an artisanal approach.


In contrast, in a thoroughgoing industrial process, it’s the other way around. Humans adapt to machines, as machines define standards and work routines.

While popular notions of artisanship emphasize its adaptation to nature—such as ambient temperatures and seasonal fluctuations in ingredients—artisans interact with many other variables as well. Participation in certain markets may necessitate the use of industrial equipment. Retailers may require bar codes on packaging, and customers who shop for the holidays in advance may prefer foods that are vacuum-packed. As argued above, traditions are adaptive. The artisans in this study adapted to accommodate many dimensions of their environment—socioeconomic and natural, as well as regulatory:

Those regulations, to me, are just like that pasture out there. If those does [female goats] go out in the north pasture and they eat wild mustard leaves all day or wild chives, I’m going to taste it in the milk. So I need to adapt.

Farmstead cheesemaker

The next section presents a more conceptual actor-network analysis of the interaction of inspectors and artisans during food safety inspections.