The production of a risk assessment is in fact the culmination of many scientific voices passing through numerous cultural spheres.
The Hazard identification stage is largely qualitative as the scientific voices concern themselves with gathering information to see if a specific exposure may have an adverse health effect of a population. There interest is in the nature, likelihood and evidence of any possible health risks (Hertz-Picciotto 1995:27-29). Yet hazard identification generally focuses on one health effect, or “endpoint”, at a time and ignores other possible harms (Israel 1995:483). Furthermore, hazards are studied under laboratory conditions which can be very different then that which the community is experiencing (Novotny 1998:141).
With a dose-response assessment, the scientific voices examine published quantitative data and prescribe risk as a relation between dose and adverse health effects. A dose-response analysis is often based on animal experimentations that are then extrapolated to humans. This is accepted despite the differences in reactions to chemicals between animals and humans (Israel 1995:483). Furthermore, these extrapolations often use healthy white males as its standard (Israel 1995:486).
In the exposure assessment stage, the scientific voices seek to identify the specific agent and determine both the duration and route of the exposure. In terms of assessing exposure, risk assessments generally overlook the multiple exposures and increasing vulnerability that many contaminated communities face (Novotny 1998). The risk for exposure may be increased due to issues such as race, class or gender. These are components of the resident’s reality that are often neglected when considering exposure. They result in barriers, inequalities, and disenfranchisements due to the biases and power differentials inherent in the system where they reside. In other words, “Structural violence is one way of describing social arrangements that put individuals and populations in harm’s way… The arrangements are structural because they are embedded in the political and economic organization of our social world; they are violent because they cause injury to people … neither culture nor pure individual will is at fault; rather, historically given (and often economically driven) processes and forces conspire to constrain individual agency” (Farmer 2006).
At the final stage, the risk characterization component of risk assessment construction allows the scientific voices to combine the data learned through the previous steps into a ratio that predicts a population’s risk to a potential exposure. The inaccuracies and misgivings of the previous steps are pulled forward to present an “objective” account that is used to set governmental policy and regulations. The EPA acknowledges that this risk value has “uncertainty spanning perhaps an order of magnitude,” yet risk management decisions are “made as if these risks are precise point estimates” (Felter and Dourson 1998:246).
Despite the uncertainties, assumptions and biases, science is valued for its ability to deliver an objective resolution to issues concerning environmental risk assessments. Moreover, it is a uni-directional process with information flowing from scientist to passive recipient (Martin 1994). Yet science is far from being objective and is instead embedded in power relations and subjective interests (Brulle and Pellow 2006:103). In their classic cultural analysis of risk, Douglas and Wildavsky said, “there is a delusion that assigning probabilities is a value-free exercise” (Douglas and Wildansky 1982:71).
Somewhere in the production of this knowledge, it is forgotten that risk assessment estimates are inherently flawed and imprecise (Douglas and Wildansky 1982:71). This may have much to do with risk communication. The public wants to know what their risk would be concerning a particular hazard with certainty and in succinct terms, rather than being explained through a quantitative process. It is also due to the risk scientists themselves who