Learning to Become an FSC Auditor: Objectivity, Interpretation, and Mastery

This paper aims to open the black box of auditing for the Forest Stewardship Council (FSC) forest management standard. Specifically, we delve into the early steps of becoming an FSC auditor by examining two auditor training sessions in northern Europe. Using a mix of participant observation and unstructured interviews, the paper subjects the trainings to a dramaturgical analysis that focuses on the ways in which objectivity was performed and on how it was taught to be performed. Alongside being an exploratory piece on FSC auditor training, this article highlights how objectivity and subjectivity are co-supportive components. Instead of being something to shy away from, auditors are implicitly taught the values of auditing even if they compromise the objective claims of the auditing process. Furthermore, the paper establishes that both interpretive and objective aspects are necessary, and that to compromise either is to diminish the capacity of the audit process.


Introduction
The first image of an auditor is that of an outsider, an interloper who swoops in to scrutinise and criticise. It is an image of ledgers, numbers, checklists, and cold, hard facts. It is also an image of faceless devotees busy ferreting away until the truth is finally revealed. These images are not so easily aligned with the daily practices of auditors as they try to make sense of the world; to reconcile what is observed empirically against the audit standards that they are required to apply. Faced with the interactions needed to compress the intricacies of life into easy-to-comprehend checklists for reporting in order to feed society's hunger for information, auditors face a daunting task.
An audit, according to Domingues et al. (2011: 1) is a "systematic, independent and documented process for obtaining… evidence and evaluating it objectively to determine the extent to which the audit criteria are fulfilled". This meaning of auditing is becoming increasingly important in a wide range of practices including business management, public policy development and environmental governance activities. The rise of auditing can be seen as part of a wider trend of what has been called informational governance (Gupta and Mason, 2014;Mol, 2008). As captured by notions such as evidence-based policy or new public management, the legitimacy of policies, including certification schemes, is seen to depend on the systematic evaluation of effectiveness, which requires the systematic collection of standardized information (Turnhout et al., 2014). Simultaneously, demand for information about how that information is produced has also risen, leading to the so-called 'tyranny of transparency' (Strathern, 2000).
In the environmental arena, this need for information has manifested in the form of third party certification audits, which have borrowed practices and language from financial accounting (Konefal and Hatanaka, 2011;Power, 1997) The Forest Stewardship Council (FSC) is one of the most prominent examples. Its logo, the 'ticktree' , is near-ubiquitous and can be found on many paper or wood products. FSC has been studied extensively from multiple perspectives. Some authors have focused on the quantitative assessment of the impact of FSC (Auld et al., 2008;Ebeling and Yasué, 2009;Moore et al., 2012), resulting in descriptive accounts of how FSC is effectuating change globally. Other literature has focused on critical examination of the principles behind FSC and FSC auditing (McDermott, 2012;Auld and Bull, 2003;Arts and Buizer, 2009). Finally, a small group has looked directly at the practical implementation of the FSC system in the field (Eden, 2008;Maletz and Tysiachniouk, 2009).
This article draws inspiration from the final group in its focus on the contextualised practice of auditing. However, rather than auditing itself, our analysis addresses auditor training material and how auditors are trained. Analysing this aspect of auditing allows us not only to examine how underlying principles and standards are presented to new auditors, but also to better understand how prospective auditors are taught to apply these principles. As with all educational or training activities, examining underlying elements is an important part of understanding the process of becoming an expert. Particular attention is paid to the importance of objectivity in the repertoire of an auditor, and by extension, the role of what is considered subjective. In so doing, we follow along the lines of Power (2003) who, drawing on the work of Bourdieu, rejects the assertion that the seemingly well-formed concept of objectivity signifies anything at all. Instead, we conclude that while the explicitly trained and well-practiced aspects of auditing are important for the overt performance of objectivity, the illformed, ill-trained aspects of the FSC audit are equally important for creating meaningful audits and performing objectivity. It is important that this is recognized in view of the increasing siginificance attached to auditing as an indispensible part of legitimate and effective environmental governance. Furthermore, it is in the interest of improving audits for this to be recognized if the principles of audits and auditing are to be prioritized in society. Power (2000) questioned the prominence of auditing in society outside of the UK, but one only needs to look at the state of FSC's certificates to see that it is a global phenomena. In 2017 there were more than 1500 valid FSC forest management certificate and 33000 valid FSC chain of custody certificates globally (FSC, 2017), each one representing a yearly surveilence audit and quinquennial recertification audit. These audits are perfomed by approximately 40 organizations (ASI, 2017) of varying sizes from a few individuals to major multi-national organizations. Considering the magnitude of FSC certification as an industry, and the importance of its values such as transparency, accountability, and objectivity in environmental governance (Gupta and Mason, 2014;Mol, 2008;Power, 1997), it is crucial to examine the explicit and implicit aspects of auditing as auditors themselves learn them.
By presenting an analysis of two FSC auditor trainings in 2013 and 2014, this article contributes to the understanding of auditing as a practice. Before presenting our findings, we first provide an explanation of our methodological and conceptual approaches, followed by a more detailed introduction to the FSC auditing system.

Expertise and the performance of objectivity
The definition of auditing offered earlier emphasizes the term 'objectively'. Objectivity indicates a rigid system by which auditors gather and analyse data. These data are subsequently confronted with the standards in order to determine whether they are evidence of conformity or non-conformity. However, as in other processes of knowledge production, there is a social dimension to auditing. Porter (1995: 35) explains "strict rules are almost indispensable unless those gathering the numbers are themselves very well socialized in the craft." This suggests that following rules is important for novices, while those with more experience are apparently able to transcend the intellectual pursuit of 'objective thinking' to the point of being intuitively a master of a craft (Yanow, 2015). Mastery of a craft means that expertise has reached a level where expert judgement no longer requires deliberation or justification. This is derived from Dreyfus and Dreyfus (2005: 779) who argue that "intuitive judgment is the hallmark of expertise". Applied to auditing, Pentland (1993) describes how seasoned financial auditors conclude their activities only after preforming the 'ritual' of becoming 'comfortable' with the numbers, which is far from an objective measure. Collins and Evans (2007) characterise expertise, working their way through their 'periodic table of expertises' from those skills which all members of a group share (e.g. common language) to the highest level of expertise (e.g. competently doing the activity). In between these extremes is the production and acquisition of knowledge or facts without context or 'practical competence' (incidentally, this is precisely the outcome of the training we analyse, but this will become evident later). For this so-called interactional expertise, no amount of if-then statements can capture the knowledge necessary. It begins to transcend the common language, relying entirely on the language of the expert-group. Essentially, one can 'talk the talk' even if one cannot 'walk the walk' . These descriptions of expertise are subject to limitations when the knowledge is subjected to high-risk, low margin of error situations (Turner, 2010). Types of expertise that demand fast adjustments with little room (spatially or temporally) for error correction do not have the luxury to disguise the errors and must sometimes instead acknowledge their "interpretive flexibility" (Turner, 2010: 250). When there is a failure in the knowledge of the expertise, the expert is forced to improvise, drawing on the baser elements of Collins and Evans' periodic table. This includes non-deliberative, non-knowledge -based elements. Kotzee's (2014) review of the literature of expertise criticized the 'fluency' narrative, which can be seen in the references to 'intuition' and baser skills in the texts above. The author's concern was that in-the-field experience is seen as the only real teacher of expertise. As such, teaching has become concerned with the performance of expertise rather than the outcome. The author also objected to the idea that the so-called tacit knowledge of expertise was tacit only because physical constraints prevent effective characterization of fast-paced or complex activities, and therefore no attempt is made to teach it. The author called for a 'social realism' perspective on expertise in which academics, educators, and experts themselves "see expertise as a real and objective ability to accomplish something in the world that enables the expert of give advice to others…" (Kotzee, 2014: 176). As such, it is important to keep in mind the "matter of how the expert… acts or… their own interpretations of their work. " (Kotzee, 2014: 176). Kotzee (2014: 176) goes on to critique the training of experts resulting in a preoccupation with the appearance of being an expert, avoiding valuing "real and objective" abilities. Our article follows this line of reasoning to a point but draws on the understanding of expertise as both entirely socially constructed and also absolutely real it its effects (also see Eyal and Buchholz, 2010). This point is made in our concluding remarks. Bal et al. (2002) offer further insight into how expertise, despite its constructedness, can assume authority and produce effects. They have outlined a central tension in the authority of science; science presents itself as authoritative and objective; however, it can only be so because behind the scenes the actors are able to effectively navigate and negotiate what it means to be authoritative and objective. Gilbert and Mulkay's (1984) analysis of scientists' discourse touches on a similar phenomenon. They show how scientists make use of two distinct repertoires to establish and justify their authority. The first repertoire is the empiricist repertoire. This repertoire is most common in the formal literature. It portrays scientific results or facts as objective, "as following unproblematically and inescapably from the empirical characteristics of an impersonal natural world" (Gilbert and Mulkay, 1984: 56). The second repertoire is the contingent repertoire and is common in informal talk among scientists. This repertoire is in opposition to the empiricist repertoire because it emphasizes the importance of context, personal inclination, and judgement. While there is an obvious tension between these two repertoires, scientists are apparently able to draw upon both repertoires. Examining this tension as it takes place in auditing will result in a better understanding of the cultural normsboth formal and informal -that are embedded in auditing (Power, 1995).
Thus, what we see emerging here is a paradox of expertise: on one hand, experts need to follow strict rules and they need to be objective, while on the other hand, expertise is about mastering a craft, which involves subjective feeling and intuition. Furthermore, there are times when pure, objective knowledge either does not suffice, or is not possible. Objectivity and subjectivity must be aligned through expertise. Conceptualizing the presentation of the trapping of expertise, authority and objectivity, as performance allows us to make sense of this paradox. Goffman (1959) presented the concept of performance as a way to explain how and why people behave in certain, largely consistent, ways when confronted with repeated or habitual, self-contained events. According to Goffman, people play specific roles, which encompass all the little details that are expected when the proper situation is presented. These roles are self-contained and rigid since breaking character causes a failure in the performance. To ensure a successful performance, the actors must be aware of, and control the 'frontstage' and 'backstage' of a performance. As we will show, the frontstage is the formal and public part of the activity. Here objectivity is emphasized as in Gilbert and Mulkay's (1984) empiricist repertoire. While in the backstage, behind the scenes, all kinds of activities and interactions take place which make the frontstage performance possible, but which in themselves are not part of, and may even seem incompatible with, that performance. It is here that the contingent repertoire abounds. Hilgartner (2000) expanded Goffman's dramaturgical perspective to include documents and reports, revealing how documents can obscure backstage elements, in particular the messiness of scientific processes, and either implicitly or explicitly bring authoritygranting elements, such as claimed domains of expertise and partnered organisations, to the frontstage performance. Applied to the paradox of expertise identified earlier, this suggests that the successful frontstage performance of objectivity depends on the intuitions and values that are part of the mastery of the craft that needs to be kept backstage.
How then is objectivity performed in auditing? We draw on Douglas (2004) to further explore the different meanings that are associated with the notion of objectivity. Douglas argues that objectivity can be interpreted in three interrelated ways, which she calls 'modes of objectivity' .
1. Object-oriented objectivity: how knowledgecreating actors "get at" objects in the world. 2. Value-oriented objectivity: values that are upheld by observers in order to know something objectively, such as detachment, disinterestedness or neutrality. 3. Process-oriented objectivity: the methods and processes that are deemed necessary in order to produce objective knowledge.
We assert that each of these interrelated modes emphasizes the performative dimension of objectivity: in order to demonstrate objectivity, actors must show that they have the right values and have followed appropriate procedures, and they must convince outsiders that in doing so, the knowledge produced refers to the object, not to the subject. In most uses of the term, subjectivity is the opposite of objectivity: if knowledge is subjective, it cannot be objective. According to Douglas (2004), the negative connotation of subjectivity is misplaced. The knowing subject cannot be erased, but forms a component of the process of producing knowledge and objectivity. We have also seen this in our earlier discussion of mastering the craft and of the importance of the backstage for the successful performance of objectivity.
As Porter (1995) has shown, objectivity is evoked mostly in fields that face outside scrutiny. This explains why it is important in auditing practices. Auditing itself is a form of outside scrutiny, and in order to be authoritative both towards those who are scrutinized and those who do the scrutinizing, it must be held to the highest standards. Keeping in mind the paradox of expertise -the fact that backstage processes of subjective interpretation are vital for the successful frontstage performance of objectivityhow are prospective auditors trained to perform objective audits and become masters of the craft of auditing? How do they learn to navigate the paradox of auditing and 'do' objectivity as well as interpretation? The element of training adds an important dimension to this because the auditor training is itself a performance in which the teachers must convince their pupils that they are experts in their field while at the same time they must teach initiates how they do auditing. To use a well-known expression: they must not only convince the students to eat the sausage, but also show them how it is made.

Analysing FSC audit practices
Our analysis is based on materials from two FSC auditor training courses, which took place in northern Europe in 2013 and 2014. Information was gathered via active participant observation by the first author who enrolled in the courses as a trainee. The participants were informed that the first author was there in an academic capac-ity and would be writing a paper based on the course contents and trainee actions and reactions, and they all consented verbally in the presence of the trainers. The trainings were held in a hotel conference room with a U-shaped arrangement of tables and chairs facing an open space that included a projector screen (it was the same location, setup, and content both years). The trainings were attended by an international group of 12 and 13 auditors-in-training. Over the course of the following days, the majority of the time was spent listening to and watching the trainers present slides, each reproduced faithfully in the training manual, while they responded to questions from the trainees. This was interrupted by exercises where the trainees were told to divide into smaller groups in order to practice the content that had been covered recently, or the entire group was asked to respond to items displayed on the projector. The courses culminated in a multiple-choice and open-response exam covering all the content. If successfully passed, the trainee received a certificate from the training organisation certifying successful completion of the course. The lead author of this article participated in all activities as well as in informal socialization during breaks and evenings.
During the training, the lead author's focus alternated between cataloguing and critiquing the substantive details of the training material, and observing the reactions, questions, and conversations of the trainers and trainees. Data was collected by means of note taking and audio recording when possible. The result was approximately 25 type-written pages of field notes per training course (divided into 'direct observations' and 'interpretations'), compiled daily and revised over the weeks following the courses. These notes included in-situ observations, as well as post-hoc interpretations of training manual contents (documents, PowerPoint slides, and images) and utterances during the training (both from notes and from transcribed audio recordings).
Analysing this data proved challenging due to its heterogeneous nature. Following Law's (2009) approach to iterative interpretive analysis of mixed content , interpretive analysis of the content was performed during the training, at night following each training day, and 'out of the field' in the weeks and months following the course. By reading and re-reading the notes and training manual, it was possible to classify them according to emergent themes and patterns. Particular attention was paid to key buzzwords, including those referring to the modes of objectivity that were used, and the meanings associated with them. This was supported by considering the use of images that were paired with particular phrases, and the physical reactions of the trainers and trainees. Furthermore, due to the participative nature of data collection, the researcher served as the primary tool of measurement, and so personal reactions to the course content served as an input to understanding the training.
The findings presented below are the result of the iterative process described in this section, as informed by the theoretical perspective above. The first part of the findings is structured on the basis of the modes of objectivity in order to highlight the multiplicity of ways in which objectivity is performed during the training. The second part of the analysis focuses on how the trainers perform for their students that they are masters and teach them how to become the same. We end by discussing how the tension between performing objectivity and performing mastery is mitigated in the performance itself through auditing/auditor values.

The Forest Stewardship Council
The basis for FSC's Forest Management certification system is its ten principles of responsible forest management. These are the same regardless of the location or type of forest to which they are applied and are established and revised periodically by FSC. These principles define what FSC regards as responsible forest management practices and are operationalized by a set of criteria and indicators. The criteria serve as the firstorder test if the principle has been met, and are generic, applicable to most locations. Indicators, on the other hand, are country-specific and represent the specific management elements that must be monitored in order to determine if the criteria, and therefore the principles, are fulfilled.
The first year, and every fifth year thereafter, that a forest management entity wishes to be certified, a certification body (CB) is hired to perform an audit for all the principles, criteria, and indicators. In the interim years, surveillance audits focus on a selection of principles and criteria and associated indicators, purposefully limiting the scope of the audit. During the audit proceedings, auditors are expected to carefully examine the forest management procedures and processes to determine if they conform to the principles and criteria, or if they are non-conformant. In the case of non-conformance, the auditor determines if this is a major or a minor non-conformance and reports it as such. In the case of a major non-conformance, the managers have 6 months to respond satisfactorily to the findings, while a minor non-conformance has a year to respond. If the responses do not come, or if they are inadequate, the certificate is suspended.
Each audit results in a report, some of which is made available on the FSC International website, and some of which is not made public. The report details the nature of the management, what species are present, the history of the management, and the products produces (from round logs to finished goods to non-timber forest products), as well as a catalogue of non-conformance reports (NCRs).
Auditors are required to meet certain requirements before they can be accredited as FSC auditors. One such requirement is a minimum number of hours of training (FSC, 2016). The list of topics to be covered by training is extensive but not exhaustive, detailed while remaining vague. Individual auditing firms have their own training requirements that auditors must meet as well.
To gain entry to the world of FSC auditors, one must be prepared to immerse oneself in training material.

Performing objectivity
The trainers introduced the notion of objectivity early in the training. After an opening title slide featuring auditors talking to a man in a piece of forestry machinery, the trainers presented a slide titled "What is auditing?" in large, bold letters. Their definition, "Objective and independent evaluation of conformance to specified and agreed requirements", stood next to an image of an evenly-balanced golden scale. Verbally, the instructor added that it is important to remember that the audit is against something specific, not what the auditors think is right or wrong. A second definition of auditing was presented shortly after the first. It was the International Organization for Standards (ISO) definition of an audit, characterising it as a "systematic, independent and documented process for obtaining audit evidence and evaluating it objectively to determine the extent to which the audit criteria are… fulfilled". There was no discussion of what was meant by the elements of this definition, just nods and note-taking by the trainees.
The term objectivity appeared many times more during the first half of the training, being expressly named 13 times, but slowly disappeared as the training progressed. However, given its prominence in the opening definitions, it is of paramount importance to the role of an auditor. While it was never explicitly stated what is meant by 'objective' , we can get a sense by noting the connotations associated with it. The image of a balanced scale connoted fairness, justice, and impartiality. It also suggests the removal of the human contextual element, replacing it with technology that has an expressly designed purpose of measuring accurately and precisely.
As its position is so important, many elements of the training explicitly or implicitly communicate how the role element of 'objectivity' should be performed. We examine the aspect of objectivity using Douglas' (2004) three-part frame of the term: Object-oriented objectivity, values-oriented objectivity, and procedurally-oriented objectivity. The details of the 'modes of objectivity' (Douglas, 2004) will be discussed in the following sections.

Object-oriented objectivity
In order to "get at" the true, objective facts of the audit, the trainers put forward a model of action that, if followed, would lead auditors to be certain of what is observed. 'Triangulation' was named as the proper method for "gathering audit evidence". The trainees were told that if something is observed, an auditor must seek out "all" the evidence, either supporting or contrary, through documents and interviews. If pursuing a detail through multiple lines of inquiry leads to the same evidentiary result, the auditor can be sure this is the true, objective reality. Figure 1 reproduces how this principle was presented during the training.
The trainers went on to define "audit evidence" as "records, statements of fact, or other verifiable information". Therefore, in order for something to be considered evidence at all, it must be documented and possible to check it against something else. During the training, it was stated that it is preferable to check a possible piece of evidence against another type of evidence but checking against the same type of evidence is also acceptable if circumstances prevent crosschecking.
Cook et al.

Science & Technology Studies XX(X)
This form of objectivity served as the foundation for the whole training. Nothing could be stated without the express reference to multiple sources of evidence, or at least with the offhand comment that in a real audit, we auditorsto-be would look for something to support or refute our findings in other sources. The objects of the audit, therefore, are not necessarily the forests themselves, but rather the management artefacts around the forests. By ensuring multiple avenues of evidence, the foundational objectivity is supposedly ensured.

Value-oriented objectivity
The trainers dedicated an entire section of the training and its manual to explicitly laying out the aspects of proper auditor behaviour and attitude. This was led by the assertion that the purpose of an audit is "to find POSITIVE evidence of conformance" (emphasis theirs). This means that rather than focusing on finding evidence of what was going wrong, we were told to focus on evidence of what was going right. In order to get "the real" evidence, auditors need to maintain an "open" frame of mind. The term 'open' was the second most repeated buzzword of the training, and usually referenced how an auditor or audit process should be. Valuing 'openness' ensures an auditor is willing to investigate all sources of possible evidence to "get all the information" (according to the method of triangulation) before deciding about conformance or nonconformance. Adhering to the value of openness allows auditors to focus on what is observable rather than on what is valued by the auditees; this allows the auditor to not be blinded to unpleasant or unexpected findings.
The trainers underlined this by showing an overarching communication flowchart juxtaposing "openness" with "criticism". Figure 2 shows how openness is inherently a virtue that leads to a positive outcome, while not being open inexorably leads down the path to withdrawal from communication. Thus, auditors should promote the value of openness because that ultimately leads to communication of information, which allows for triangulation regardless of the values influencing the situation.
The topic was complicated, however, by the idea of 'neutrality' which the trainers introduced later in the training. The trainees were instructed to "view without prejudice" the situations during the audit, but to nevertheless come to understand why we see what we see. Therefore, the valueladen aspect of human-managed systems is given weight, but supposedly not allowed to influence the findings of the auditor. Auditors should have the "ability to understand the… situation," although there is no need to take sides on the values at stake. Thus, an auditor can transcend these values by remaining neutral and thereby maintain the objectivity of the findings.

Process-oriented objectivity
A large portion of the training was dedicated to the procedures of an audit; it focused on engag-

Criticism
Defensiveness Arguments Closure/ withdrawal Openness Responsiveness Cooperation ing the right processes to attain objectivity. These processes would ensure that, as summed up in an offhand comment by the trainer: "anybody would come to the same conclusion". In theory, if an audit were performed twice, and the same procedures followed, the same conclusions would be reached regardless of the auditors present. However, in the same breath, this situation was problematized when the trainers introduce the idea of the "unforeseen situation". At this point in the training, the trainers were vague about what such a situation may be, but we experienced simulated "unforeseen situations" in the exercises discussed later. The addendum of the unforeseen suggests that each audit is unique, and following standardized procedures is an attempt to control a chaotic system in order to claim a form of objectivity.
During the first day of the training, the trainers spent time describing the procedure of an audit, which included establishing the shared vocabulary of an audit. To someone unfamiliar with the audit process, terms like 'CB' would be meaningless, and indeed, most of the trainees seemed to know what it meant, as the lead author was the only one who felt the need to ask. Having, and effectively utilizing, a shared and codified vocabulary can serve two purposes: First, it identifies to others in the auditor role that you are also in the role of auditor, and it also eliminates imprecise, humanising language. Throughout the training, the trainers, and progressively the trainees, continued using terms like 'CB' , 'surveillance audit' , and 'NCR' (non-conformance report) consistently. By employing such codified, expert language, auditors can supposedly be more precise and can attempt to eliminate personal idiosyncrasies.
Due to the nature of the training, field observation procedures were not very well covered, as all material had to be presented to the trainees in a classroom setting. However, the topics of interviewing and communication had a well-defined set of recommendations, almost to the point of being a script. The value of openness identified earlier came strongly to the fore here. Specifically, it was considered important to avoid closed and aggressive forms of what they called "shoot and reload" communication. Rather, to promote open communication, the trainees were instructed to use "open questions" to promote "conversation" instead of "interrogation". This included a list of stock-phrases and questions to be used ("Tell me how…", "what is the procedure…", "why have you…").
Following this instruction, we performed an exercise to test our "interview skills". We were playing the role of auditors who had a short time to interview a health and safety officer and a chainsaw operator (played by the trainers). Based on the way the interview progressed, the trainers offered critique if we interrupted or did not ask sufficiently 'open' questions. The purpose of the exercise, it was explained, is to allow those being interviewed the chance to give you all the information so you can come the correct conclusion.
Alongside the information collecting, the trainers also addressed recording and reporting by introducing the NCRs. NCRs were described as the "mechanism for requiring the auditee to take action to meet the standard". It was previously established in the training that auditors should communicate non-conformities in an "objective and diplomatic way" and "be open for additional information". Therefore, even concluding that there is a non-conformance may change in the face of additional evidence. To present the NCR, the trainers showed an example NCR form with pre-determined fields for describing the non-conformance, "corrective action request" (or 'CAR'), and timeline for conformance among other fields. The result is an inflexible, but subjectively descriptive way to frame the observations made during the audit.
During the substantive teaching of the training, all three modes of objectivity emerged as being important to the success of an audit. No type of objectivity was given precedent over the others, as process-oriented objectivity mingled equally with statements on value-or object-oriented objectivity. Through this explicit or implicit discussion of objectivity, we can see how each type of objectivity is trained to be performed during an audit, and how they are co-supportive.
In the following section we examine how objectivity might be done in the chaos of the field. We discuss several exercises which served to simulate the chaos of a "real audit". While the application of procedures described previously was encouraged, in the end, the trainees needed to step away from the cold application of knowledge to make it practical. As we will describe in the following section, trainees were expected to interpret the situation using heretofore undefined or poorly defined criteria.

Performing interpretation
During the training, objectivity was actively named and encouraged as something to promote and aspire to. In the previous section we discussed the most prominent ways in which objectivity was invoked. There emerged another aspect of being an auditor that did not follow the zealous application of objectivity, and was, mostly implicitly, termed interpretation.
"Interpretation" was mentioned explicitly only once during the training. In the section regarding auditor behaviour (discussed previously), interpretation appears in a communication model explaining how auditors should strive to understand what is truly meant by the auditee and understand that the auditee must do the same. Figure 3 reproduces this image. In this context, interpretation is something that is actively done to decode encoded statements. The trainers fully acknowledged that despite having technical know-how and information about the management, it is still possible to not get at the real "reality" of the situation. However, previously, the trainers had stated that there is "no one reality", and that it is an auditor's job to discern which reality presented by various actors is most valid, or the really real reality.
The trainers identified this as the "ability to understand… the situation" but were not able to characterise it beyond the methods, procedures, and facts described in the previous section. In order to actually "understand… the situation", we were told to pay attention to auditee tone of voice, body language, and what was not said as much as what was said. The idiosyncratic nature of these elements meant that the trainers were only able to speak in broad terms, and it was up to the trainee to determine when there was "enough information" to judge the situation. For example, the trainers explained that if a worker seems reluctant to speak, failing to make eye contact or speaking softly, it is a good idea to speak to them on their own later as they might be hiding something. We asked about how to consistently interpret these elements, but the trainers were not able to give meaningful answers. The term "interpretation" was clearly not acceptable for this type of analytical work. Nevertheless, we were being asked to observe and interpret the actions and utterances of the auditees.
One set of exercises where interpretation was practiced was roleplaying exercises. We had access to mock forest management plans and were expected to effectively and efficiently divide the work to assess conformity. However, they also included "unforeseen situations". For example, a gregarious manager who wanted to have a celebratory drink, and a business partner who was quiet when in the presence of the other partner, but talkative alone. If we, the auditors, were not sufficiently strict with the former or did not interpret the downcast eyes of the latter properly, the trainers would critique our interpretive skills. Figure 3. The only official use of the term "interpretation", presented in the training in a highly simplified communication model. The trainers acknowledged that these situations were artificial and contrived in nature, but they were meant to simulate the chaos of a real audit. If we did not follow the proper interviewing guidelines, or failed to utilise the approved language properly, our interpretations were considered fallacious. In particular, the role of properly distinguishing between friendliness and deception, properly understanding body language, and tone of voice were practiced. Each team (3-4 people) was able to watch the others perform their interviews and offer their own critiques. These exercises still relied on executing the 'audit procedures' properly but required the trainees to interpret and react on their feet without the benefit of conscious, rational deliberation.

Science & Technology Studies XX(X)
This was continued with a photo identification exercise. In the evening of the second day, the trainers presented a series of slides containing various scenes around a forest management unit.
We were asked to look at each picture for a few seconds at a time and determine if there were any non-conformances. If there was a problem, we were expected to identify which part of the FSC standard was being violated. The results were mixed. Some non-conformances were missed, but more often, non-conformances were identified when in fact there were not any. The lead author made at least 4 incorrect identifications of non-conformances when, in fact, there were no problems. Following the exercise, there was discussion where trainees were asked to reveal their judgement of the photos, and to justify their findings. While some were definitively ruled as correct or incorrect by the trainers, more than once the trainer conceded that there was missing information or room for classifying it either minor or major non-conformances. When put in a situation that needed snap-decisions, we were expected to rapidly interpret the situation and come to a conclusion utilising the details learned during the course.
One of the trainers brought up a clear example of interpretation in a situation where a snapdecision was required. To illustrate the point of knowing how to "split auditors appropriately", dividing the audit team to cover more ground in an efficient manner, the trainer told an anecdote wherein, on a "hunch", he thought there was a need to investigate a certain area of the forest. There he found evidence that the managers were hiding evidence of all the injured workers from the auditors. When pressed on how he knew to look for this, the trainer gave a small hand wave, and an answer along the lines of "I just knew" based on all the glances, winks, nods, utterances, and silences he witnessed during the audit.
During the training, it became increasingly clear that interpretation is demanded of auditors in almost all aspects of their work: in assessing management plans, in exploring situations, in understanding the behaviour and responses of auditees and in making judgements about conformance or non-conformance. It was also clear that, despite the emphasis that was put on procedures, there can be no procedure for interpretation; it can only be based on experience and intuition in the moment.
The front staging of interpretation during the training, like in the example of auditors acting on a hunch, triggered two simultaneous responses. On one hand, it was seen as evidence of the authority of the trainers. At the same time though, it posed a problem for the trainees because they did not know how, nor could they be taught to do this themselves and maintain objectivity. The trainer was unable to communicate the imponderable details that led up to this insight; he could not provide an answer that could help the trainees learn how to do this themselves. Nevertheless, through the gesture of the shrug, he showed that he recognized the problem but that it did not bother him. Forgoing deliberation helped reinforce the necessity of becoming an authority and satisfying the audience.
This situation was echoed in another manner during the training. Over the course of the training, approximately six times when a trainee would ask a question or pose a hypothetical situation that would address a tension between the seemingly straightforward procedures for auditing and the need for interpretation, there would be a moment of pause, after which the trainers responded in a similar way as described above. They would try and offer an explanation and while the answer was relayed, a hand would be waved through the air. This gesture specifically became attached to the phrase "we all know how it is" when it was repeated three times in short succession by two trainers and a trainee when describing hypothetical forest management practices.
As we will argue in the next section, these gestures are an important part of the trainers' performance. They are where objectivity and interpretation become entwined in the performance of mastery. Utilising Douglas' (2004) modes of objectivity, we have demonstrated how FSC auditors are taught to perform objectivity in audits by trying to get at the object, following the right procedures, and adhering to the right values. In agreement with her analysis, we find that most forms of objectivity are encouraged, though not necessarily equally, in most situations. We have also seen that the subjective, in the guise of "interpretation", and the objective overlap in many cases. In fact, interpretation is necessary for the successful performance of objective audits. Thus, subjectivity is not the antithesis of objectivity, but rather a vital support of it.

Performing mastery
In the training, they often belonged to different settings. Objectivity was most prominent in the more substantive parts of the training where content was taught explicitly. Interpretation, on the other hand, could not be taught in the same way as the elements of objectivity. It was taught as something done in the mind of the auditors, or within a team of auditors based on the information gathered while employing the modes of objectivity. It was highlighted in stories of the trainers and it was practiced in exercises. Trainers also assessed and critiqued the performance of the trainees in both aspects. They assessed our level of knowledge (for example, identifying the correct part of the standard to use), but also less tangible skills (for example, spending too much time complimenting good management or not asking sufficiently open questions).
Objectivity and interpretation did not clash but smoothly alternated and aligned for the most part of the training. Generally, interpretation was implicitly recognized as important and non-threatening. One way in which this was done was by taking the FSC standards and criteria as the immutable starting point of interpretation. While it has been said that standards are not in fact stable or immutable (Eden, 2008), this is not how they were presented during the training. By initially referencing something that neither the auditors nor the auditees had any hand in creating, something that cannot be changed by the involved parties, the auditors are able to claim objectivity. Therefore, it is never the auditor who is saying if a management practice is (in) sufficient, it is the standard saying it. Although interpretation is seen as part of process, specifically when the findings of the audit are tested against the standards, this interpretative part quickly disappears from view when the findings become objective evidence of conformance or non-conformance.
However, in some instances, interpretation could not be hidden or denied. In the example of the "hunch" leading to discovering hidden injuries, the trainer revealed how he had to make a judgement on whether to follow his interpretation of the situation (his "hunch"). He decided to "go with his gut" and this led him to certain results that after proper documentation could be used as objective evidence of non-conformance. While the judgement of non-conformance could be justified by referring to the evidence and their testing against the standards, it does not meet the expected burden of object-oriented objectivity of auditing. It would be entirely possible for another auditor to not make this decision and miss the evidence. When asked during the training, the trainer was unable to explain what exactly made him suspicious. In this story, and in other similar anecdotes that were shared during the training, the tension between objectivity and interpretation manifests itself on stage. While in most cases, interpretation can be kept hidden in the backstage, in the context of the training where interpretation is taught and put on stage, the trainers must find a way to mitigate or transcend this tension. This was not just a matter of flexibly, manoeuvring between contrasting repertoires as Gilbert and Mulkay (1984) suggest. Instead, this transcendence was accomplished in bodily performance; in the shrugs, nods and handwaves.
We suggest that these gestures, combined with phrase "we all know how it is" are central to the performance of mastery. This performance of mastery illustrates Dreyfus and Dreyfus' (2005: 787) argument that the "proficient performer" has a "repertoire of situational discriminations" that allows for "immediate intuitive situational response [s]." It fulfils several functions. First, it bypasses cumbersome explanation. Thus, the successful performance of mastery is characterised by not needing to justify or explain the interpretative process. Second, it recognizes the trainer's inability to explicitly explain or justify while at the same time dismissing it as insignificant, as unthreatening to the objectivity of auditing. Third, it creates a sense of community among trainers and trainees. By shrugging or handwaving and communicating "we all know how it is", and by employing the same, codified language, the trainer conveys what Goffman (1959) calls a role secret. It signifies that trainers and trainees are all member of the auditing community while still maintaining processoriented objectivity (Douglas, 2004). Trainees could respond, and many did by nodding understandingly, confirming their membership of the community. Finally, it serves a didactic purpose: regardless of whether trainers or the trainees do in fact "know how it is", the trainees get a sense of what it might entail to perform mastery without having to fully come to terms with what is needed to be a master. The attention paid to interview and interviewee affectations, as well as the importance of auditor language and behaviours can lead an auditor in training to be a master at adopting the mannerisms of a master with the assurance that true mastery will come through practice.
This aligns very closely with the notion of interactional expertise. By progressing from strictly knowing the content of an audit as a function of promoting objectivity, to accepting the role of interpretation, to mastering the metaaspects of an audit performance, the lead author was inducted into the ranks of FSC auditors. In the following section, we discuss some key points from the analysis and conclude with broader implications for the field of forest certification auditing.

Conclusion
Objectivity was considered fundamental to auditing during the trainings. Using Douglas' (2004) modes of objectivity, we highlighted how objectivity was promoted explicitly and implicitly during the training. According to the principles of the training, however, attempting to do objectivity in the field inevitably leads to a breakdown in objectivity. The knowing subject (the auditor or trainee) is forced to inject her/ himself into the situation in order to properly interpret what is going on. From an outsider's perspective, the resulting paradox could be seen as problematic, as interpretation seems to be anathema to objectivity. However, the trainees are taught that by properly performing the audit, it is possible to simultaneously perform objectivity -establishing and maintaining an object separate from the subject -and perform interpretation. In doing the audit or practicing the doing of an audit, objectivity and interpretation become co-supportive or co-constituent. In the performance of mastery, the paradox of auditing is mitigated or transcended almost completely without conflict. While the current analysis is based on auditor trainings, where interpretation cannot be left in the backstage but must be put on stage for teaching purposes, the same co-constitutive relation between objectivity and subjectivity is expected to take place in auditing practices (Eden, 2008). Thus, our analysis has implications for our understanding of auditing and other practices of the construction of evidence in natural resource management and policy more generally.
While we should not overstate the possibility of extrapolation based on only two training sessions, there are two important factors to consider. First, the training organization is considered a preeminent external trainer on the topic of FSC. This implies that they will be the trend-setters and their training will be the entry point for many auditors globally, who will then go on to train others. Second, the training itself espoused that fundamental auditor and auditing characteristics are universal and absolute. Therefore, the insights generated by analysing auditor training will also be relevant for FSC auditing.
Our research showed that the 'fluency' model, identified by Kotzee (2014) is very much alive and well in the world of FSC auditing. As noted, the trainers pointed out how the training cannot possibly replace real-world experience and that instinct and intuition is what often gets auditors through rather than deliberation. The result is that the leg-work of the audit is hidden in the end, making it look like a charlatan's magic trick (Fuller, 2006, cited in Kotzee, 2014. For example, the trainees were instructed to "find positive evidence of conformance". This suggests that it is not the goal or purpose of an audit and auditor to find what is wrong with the management of a forest. Furthermore, it is not considered sufficient to simply not find any evidence of wrongdoing. Auditors are expected to find evidence of rightdoing, even though in the official reports there is no space for these findings. Thus, an activity with the goal to reveal and increase transparency may, in the end also conceal and increase opacity instead, as noted by Strathern (2000).
In this paper we focused on, and are critical of, the tension between objectivity and interpretation during the training course, but it is important to note how each aspect needs the other in order to function meaningfully. Eyal and Buchholz (2010) noted this in their literature review as the issue of 'interstitial domains' , where technical and non-technical (or political and apolitical) collide outside of well-defined domains or disciplines. The job of these auditors is, ultimately to make a recommendation as to whether, in sum, the auditee does or does not meet the standards set by FSC. In the end, this cannot be done without the auditors making a judgement call, although they are averse to the very idea of 'judgement' . During the training, the notion of coming to a conclusion was predicated first upon having all the information, but later upon having enough information to pass a judgement. However, there is no clear point at which enough is enough, and obviously, it is unlikely that one can ever possess all information on a given topic. Thus, except in artificially simplified cases, eventually it comes down to the auditor and his team saying, 'in sum, I think they are/aren't in compliance' . This exemplifies how the process in which observations are tested against a set of criteria and standards and become evidence involves a creative and interpretative leap that cannot

Science & Technology Studies XX(X)
be justified completely with reference to the objectivity of the observations or the standards .
Our analysis of learning to become an auditor has its foundations in the values present in the learning process. While the training itself goes to great lengths to characterise auditing and auditors as objective, and the trainers presented the values an auditor should possess to do auditing well, there was no attempt to identify them as actually value-laden, and therefore subject to the influence of the morals of the times. These values include the explicit, such as objectivity and assumed conformance, as well as the implicit, including both ontological relativism ("there is no one reality") and ontological realism (finding "the real" evidence). It seems that auditor trainings could become much more salient and instructive if the dilemmas associated with these values (how do you prove that something is positively true, how do you get to the truth while at the same time taking all representations of reality seriously) were openly discussed and deliberated, perhaps resulting in ultimately more effective audits and auditors. The more crucial point, however, is that none of these values are inherently bad or wrong, regardless of the implied or explicit position that is taught on the topic of personal values. In fact, they serve auditors in overcoming the paradox auditors face between objectivity and subjectivity.
Along these lines, Collins and Evans (2007) argued that the study of experts and expertise had become too concerned with how expertise is attributed. Kotzee (2014: 176) expanded this critique to include the trainers of experts and experts themselves: the education of experts was concerned only with the appearance or procedures of being an expert, focusing only on their own interpretations, while avoiding "real and objective" abilities to accomplish their tasks. While we agree that such navel-gazing should be avoided, it seems this critique has little value. The training of FSC auditors seems to surpass the point entirely, as it included the performance of being an auditor (values and procedures), and the impact of being an auditor (monitoring and reporting). By examining the learned values of auditors, as suggested by Collins and Evans (2007), we can see that the tension between expertise as performance and expertise as reality does not exist in the field of forest management certification auditing.
This lies close to the heart of a larger issue. Rather than dismissing audits for being subjective, apparently failing to meet the standards of objectivity, we follow Bourdieu's suggestion as presented by Power (2003) that it is time to get passed this way of thinking altogether. Why does something so seemingly trivial, that the construction of evidence is a human endeavour, often become so problematic? The value of objectivity permeates decision making regarding environmental management, and yet if that were the sole goal of such activities, we would entrust this work to purely computational systems. Instead, context, skill and personal experience play a vital role and auditors are, inevitably, interpreting, thinking, reflecting subjects. Our criticism of objectivity does not, however, imply that we abandon objectivity as an unattainable ideal and celebrate subjectivity and interpretation (also see Cook et al., 2016). Rather we suggest that it is time to recognize that this is how objectivity is done: by interpreting subjects. Pointing to the role of subjectivity, values, and interpretation in audits in that way does not diminish audits' objectivity but gives important insight into Cook et al. objectivity's production in practice, as hinted at by Power's 1995 work. It follows that auditor trainings can be more open about the values to which auditing adheres to fulfil not only Kotzee's (2014) call to be more explicit in characterizing expertise types, but also Eyal and Buchholz's (2010) problem of 'interstitial domains' . As Douglas (2009) argues, being value free in the production of knowledge is not only impossible, it is also undesirable. Rather, objectivity can in fact entail the open reflection on what would constitute the right, ultimately subjective, values to guide the production of knowledge. On an even broader level, this supports the growing demand for information on environmental governance, and furthermore, information on the information about environmental governance (Mol, 2008). This is the opportunity for auditors to lead the charge on providing grounded information for environmental governance at a time when the call for such has never been greater.