From Risk Perception to the Construction of Safety in the Profession of Mountain Guide

Our summary

This research falls within the field of safety in artisanal activities. It aims to understand and improve safety within the profession of high-mountain guiding—an archetype of an artisanal system—but its insights go beyond this case, offering a better understanding of how safety can be constructed in such systems, which remain largely underexplored.

The work involves analyzing how guides, as self-employed workers, constantly regulate their risk-taking in the face of natural hazards, while also considering other dimensions of their activity (economic, service-related, social, etc.).

The data come from four studies: 

1. interviews with high-mountain guides about their risk perception as they prepare for climbs;
2. observations of the preparation phase for high-mountain excursions;
3. observations conducted during the climbs themselves;
4. and a detailed analysis of the “risky trade-offs” observed in their activities.

Very enjoyable to read, the thesis is richly illustrated with real-life cases and direct quotes, and it contains a wealth of empirical data.

The Profession of Mountain Guide and the Various Risk-Related Challenges to Manage

In France, the profession of high-mountain guide is considered the most dangerous of all. Accident data records 4,320 incidents, including 132 fatalities, between 2003 and 2018 — an average of 8.25 fatal accidents per year among approximately 1,600 active members. This very high rate (just over 10⁻²) has remained relatively stable since the 1970s, with no significant improvement. Nevertheless, safety issues are occupying an increasingly central place within the profession. 

This paradox stems from profound changes in the guiding activity itself, which faces several types of risk — sometimes contradictory — that must be managed on a daily basis:  

Adaptation and Trade-Offs

The constant requirement to adapt to changing conditions and to make trade-offs across all these dimensions forms the foundation of the model developed in the thesis.
The proposed safety model in the thesis brings together two lines of research.

Archetype of Artisanal Systems

The profession of high-mountain guide represents an archetype of artisanal systems — systems that are not highly regulated, yet are very high-performing and intensely competitive. These systems bring together occupations in which risk-taking is a core component of the economic model, and which are characterized by the ability to face the unknown, to innovate, and to adapt to new contexts.
In such environments, individual autonomy and competition between actors often take precedence over hierarchical group organization and standardized practices.
Knowledge about these models largely builds on the work of Gaël Morel, René Amalberti, and Christine Chauvin in the 2000s on the fishing profession. Their research has become an international reference in the study of safety in artisanal systems, particularly through the introduction of the distinction between “regulated safety” and “managed safety”. 

The Field of Service Activities

The profession of high-mountain guide also falls within the realm of service activities, which are characterized by the client’s involvement at every stage of the service delivery, through co-production between professionals and recipients.
This second aspect—related to the consequences of the close coupling between the guide and their client (initial acceptance of the request, renegotiation of the “contract” resembling a commercial transaction that may or may not be possible, ongoing contextual adjustments, and the resulting risks...)—is more innovative in the context of an industrial safety model.
Antoine Girard, the author of the thesis, drew heavily on work published by his academic advisors on service relationships (see Situations de service: travailler dans l’interaction, Cerf & Falzon, 2005, PUF).

It’s worth noting that these authors and their students theoretically framed this field of verbal negotiation between clients and providers, although it was not initially focused on safety.

They distinguished: 

• service activities as a general framework,
• service situations, referring to specific work contexts,
• and service relationships, which describe the specific forms of interaction between stakeholders.

From this, they derived three levels within the service relationship:

1. the relational level, for conflict avoidance and tension management,
2. the contractual level, for sharing organizational rules and establishing a shared contract model,
3. and the transactional level, for the direct link between the task and negotiation around problem-solving methods.

Adopting this perspective means analyzing any particular service situation as a professional-client pairing operating within a specific socio-economic system. Medical practice is a typical example of high-risk service activity often cited by these authors and their students. 

What stands out is that Antoine Girard skillfully and originally combined these two approaches — regulated vs. managed safety, and service relationship theory — to make an innovative theoretical contribution to industrial safety. Bringing these two analytical frameworks together makes it possible to identify and model concepts such as boundaries, margins, zones, trade-offs, and sacrifices — key elements for mapping out the adaptive space needed to manage unforeseen events. However, these elements themselves must be anticipated and controlled to stay within acceptable limits.

The Economic Dimension

Another innovative contribution of the thesis is its in-depth examination of the economic dimension of the profession as a key variable in understanding and adjusting the risk management model. 

Most high-mountain guides work as self-employed professionals. Finding clients is far from straightforward in this freelance context, increasingly requiring reliance on intermediaries (such as agencies) and the need to set their own prices — prices that must be high enough to make a living, yet low enough to remain attractive and competitive.
In reality, these rates are often undervalued: one in five guides offers below-average pricing in an effort to stay competitive, and two out of three rarely ask for a deposit, meaning cancellations frequently occur without compensation — even in poor weather conditions. As a result, more than one in four guides (28%) report not working as much as they would like and are constantly looking for new clients.
It is common for these high-mountain guides to engage in one or more additional activities in order to make ends meet. On average, they work fewer than 100 days per year, and a high degree of mobility is required: 20% of guides report spending several months a year living away from their primary residence. Under such conditions, it becomes difficult to generate sufficient income without accepting a certain level of risky compromise — even within their family life. 

The thesis continues with a taxonomy of the different types of risks specific to high-mountain expeditions.
All of this social fragility and accident data also point to the guide’s physical condition and overall health — areas rarely addressed in the literature, yet where physical, psychological, and social wear (including in private life) are particularly pronounced, especially given the financial pressures of the job.
 

Contributions of the Thesis's Four Studies

Studies on Route Preparation

The first two studies focus on the preparation of high-mountain routes and shed particular light on how risk is perceived, how guides anticipate what can be controlled and what remains more unpredictable or difficult to manage — particularly through negotiation of the initial “contract”.
We learn that risk perception draws upon information related to the upcoming situation, the guide’s own capabilities, and the client. Among these sources are personal memories of past experiences (including past accidents), knowledge of “well-known” accidents involving others, one’s own metacognition about strengths and weaknesses, and of course, up-to-date knowledge of the specific difficulties of the planned route. These difficulties are themselves the subject of a known taxonomy among guides, including hazards such as steep slopes, unstable elements, crevasses, etc., current situational risks (weather, for instance), and finally, client-related factors.
This perception of risks shapes the preparation for the task and constructs—echoing Rasmussen’s representations of practice migration—an "acceptable situation space," the result of all these constraints (personal, local, technical, and those specific to the day’s conditions and to the client). This narrows the range of possible options that can be offered while already incorporating strategies for managing anticipated contingencies in a reasonably safe manner and discarding options deemed uncontrollable.

The findings from these two initial studies highlight an important point: client-related constraints (their request, satisfaction, ability to renegotiate) are at the top and center of all the trade-offs made. These client-related factors regulate and influence all other constraints in a cascade effect, shaping what is feasible and what compromises must be made.

The aim, of course, is to preserve the priorities of the contract as much as possible — not just the client’s request but also their safety. Behind this lies another personal and critical concern: any overly cautious decision to cancel or change plans may quickly threaten the financial viability of the profession.
In many cases, the primacy of the client’s request puts the guide in a position where managing all other constraints becomes nearly impossible. This calls for skilled negotiation — drawing upon service interaction expertise — to offer a viable “alternative” that balances satisfaction, safety, and economic feasibility.

Naturally, this preparation must later be tested against the reality of the terrain and its surprises in order to fully embody the safety model. This aligns with well-established findings in international literature on work preparation — especially studies on pre-flight planning by fighter pilots (Amalberti, 1992)—and a large body of later research on task preparation and briefings in high-risk industries, much of it inspired by Rasmussen’s work on managing and calibrating a dynamic space of acceptable situations.
These pre-adjustments to upcoming activity shape safety outcomes by anticipating risks, involving significant efforts to discard some options — especially those that are less desirable or carry high levels of uncontrollable uncertainty. The most significant form of renunciation is deciding not to proceed with the task at all—yet this is often an impossible choice, as it comes at the expense of income and long-term professional sustainability, and naturally entails a much higher economic cost.

Studies on the Actual Execution of Mountain Expeditions

The third study focuses on risk management during the actual execution of high-mountain expeditions. Antoine Girard acted as a participant-observer —  in a quasi-ethnographic approach — following his fellow guides engaged with clients on high-mountain routes, carefully noting and recording all safety decisions made by the lead guide and assessing their effectiveness. 

It was observed that in every analyzed situation, the guide neither seeks to completely eliminate risk nor simply to apply the safest possible solution.

Instead, the guide continually aims to control risks sufficiently to maintain an acceptable situation. This activity of conducting a high-mountain expedition confirms the extensive literature from the 1990s and 2000s on managing dynamic situations (Woods & Sarter, 1998, 2000; Amalberti, 1999; Sarter & Amalberti, 2000; Endsley, 2002; etc.).

A significant part of the guide’s work consists of observing and integrating relevant elements to continuously build a reliable mental model of the situation and to act so as to remain within an acceptable situation over time in response to the dynamics of reality.
An acceptable situation can be understood as one in which the continuous coupling of relevant elements that compose it generates a performance deemed sufficient by the guide. The adaptation of Mica Endsley’s model of situational awareness is particularly useful to clarify what matters and what is taken into account by the guide in identifying constraints and their evolution during work. 

• The guide heavily mobilizes knowledge of the route, their own capabilities, and those of their clients to develop plans in advance.
• Further analysis of the objectives to be met during the expedition shows that the guide aims not only to avoid accidents but also to build and preserve the clients’ resources, ensure their enjoyment, while maintaining a reasonable schedule and minimizing surprises and pressures of all kinds.

Studies on Risky Trade-offs

The fourth study focuses on the real-time management of compromises and trade-offs, extending the previous analyses.
Resources must be allocated among the various previously mentioned objectives, which can be in competition with one another, constantly requiring compromises and trade-offs.
The data show that the space of possible solutions, routes, and chosen conditions is actually quite large by playing on the combination of all constraints to offer several possible settings, which explains fairly noticeable variations in choices from one guide to another.

This fourth part specifically examines, in this context, the “risky trade-offs”. These risky trade-offs involve options that increase the risk of accidents but come with compensating advantages.
They stem from decisions not fully secured by the guide’s knowledge, more endured than truly chosen — such as decisions trapped by elapsed time and the progress of the expedition. Typically, the serial coupling of trade-off mechanisms highlights the dynamic and irreversible nature of the climb. As the expedition progresses, turning back can quickly become impossible in high mountain conditions.
All trade-offs become strongly interrelated: for example, deciding to engage with clients on a particular route means having to trade off according to only the options that this initial decision allows later; or deciding to reach a specific refuge severely limits possible options to only those routes accessible from that location.  

These risky trade-offs can be understood as processes resulting from three mechanisms:

It should be noted that many trade-offs have the effect of reducing the options available for the remainder of the climb. However, it is also observed that the guide’s judgments of acceptability, expressed retrospectively about their own past trade-offs, do not seem to depend solely on the direct estimation of immediate and objective risk-taking. Some choices of riskier options are, for example, judged as acceptable as less risky choices — or, put differently, from the guide’s perspective, risk-taking cannot be the only criterion for the acceptability of trade-offs. These trade-offs must also be analyzed with regard to their expected effects, including (and especially) for the client: protection, feasibility, but also acceptability, potential disappointment, the level of explanation required, persuasion, and ultimately billing, etc.

A kind of taxonomy of these risky trade-offs made by guides points to purposes that are all inscribed within a very broad temporal scale, taking into account a spectrum ranging:

• From managing a fleeting emotion during the climb to managing personal goals over the season or career,
• From managing a client during a technical passage to their training with a view to a future interaction,
• Or even to respecting the contractual framework of the climb to maintain credibility with a contracting agency providing work, and so forth.

Overall, the practice and mastery of these risky trade-offs involve a dual compromise centered at both micro and meso levels: 

It is important to note that these risky trade-offs made by guides do not solely stem from errors or a lack of awareness of risks, nor should they necessarily be interpreted as culpable or faulty behavior in this context.
Rather, they are the result of the dual compromise of micro- and meso-level management that the guide continuously and dynamically seeks to maintain in their professional activities.
Guides themselves are not always satisfied with these trade-offs. Lived experiences allow the progressive refinement of the compromise adjustments through experience and encounters with risk situations that are judged retrospectively as unacceptable.
This highlights a skill of trade-off management, which can be seen as a decision-making capacity based on the guide’s representation of the situation’s elements, as well as the ability to consider the positive and negative effects of these decisions at different spatio-temporal scales (individual climbs, season, career).

From a supportive perspective, it is therefore relevant to seek ways to build this skill more quickly and effectively, finding means to develop it without generating excessive risk-taking. It is understood that risk-taking also results from the desire to balance contradictory dimensions at different scales.
Of course, the model has its limits: the stronger the constraints applied to the guide’s decision-making and the greater the number of elements to be managed simultaneously, the higher the chances that the regulation will lead to a situation that is unacceptable from the accident risk viewpoint.
 

Lessons for Industrial Safety in High-Risk SMEs? 

The case of the high-mountain guiding profession illustrates the extreme safety model typical of artisanal high-risk trades, much like the historical studies on combat pilots (Amalberti, 1992) and fishermen (Morel, Amalberti, Chauvin, 2008).
This model confirms what was already known about risk management in these extreme situations — particularly the trade-offs between endogenous risks related to know-how and exogenous risks inherent to the situation — and adds three important dimensions for understanding risk adjustment that were either ignored or insufficiently considered in historical models.
These are:

• the priority given to overall customer satisfaction, which often overrides the management of objective risks in the work situation;
• the priority given to the immediate economic needs of the profession (income);
• and more broadly, the priority given to the long-term economic sustainability of the trade (the survival of the business).

Many SMEs will recognize these priorities. Such a model is very relevant for shedding light on risk management across an entire class of artisanal SMEs operating in high-risk sectors: many construction SMEs working under customer pressure on which their commercial survival depends; SMEs engaged in energy services facing the urgency of satisfying clients; maritime trades, especially fishermen; various sectors such as healthcare, logistics, petrochemicals, and other niche trades. While these are all small companies, collectively they represent hundreds of thousands of workers.
It is easy to understand that in all these professions and situations, ultra-cautious classical safety models — which often lead to giving up exposure to risk — are not relevant. The essence of these trades is precisely to take risks and to control them in order to survive. This is the key entry point for any support that can be offered: to (better) live with risk without renouncing it, otherwise the trade disappears.