Whose Words Do You Use When You Develop?

Why naming a phenomenon is the first step to understanding it, and why borrowed concepts always need contextualization

An organization decides to develop "safer space." The concept feels intuitively understandable. Its roots are multiple, but particularly central are the 1960s and 1970s American women's movement, civil rights movement, and LGBTQ+ communities, where physical and social spaces were built for conversation, peer support, and political organizing without fear of violence or discrimination (Kenney, 2001; The Roestone Collective, 2014). The concept has since become established in event, hospitality, and civil society contexts to describe environments where inappropriate behavior is addressed and where everyone has the right to feel safe.

In organizational development, one systematic problem keeps recurring: measurement begins before the phenomenon has been defined.

An Example from the Event Industry

A good example of how challenging building safer space is in practice can be found in the Finnish event industry. Slush, a startup event that gathers tens of thousands of attendees annually, has based on publicly reported information worked for years on safer space questions: a Code of Conduct, doubling the number of safety personnel, harassment liaison persons, layout changes, and many other measures. This is not a story of failure but the opposite: Slush is one of the few that does this work publicly and as an ongoing process, and that is precisely why their experience is instructive. The example shows that building safer space is a continuous process in which one must keep asking what it means in one's own context, how it is recognized, and how it can be influenced. Within a single organization, perceptions of safer space often differ significantly depending on the situation, role, and background of the person involved.

Now imagine that some other organization decides to launch its own "safer space" development initiative. Training begins, surveys are sent out, conversations are held. Everyone seems to understand what it is about. And yet six months later, participants have very different ideas about what safer space actually meant and what it was supposed to produce.

The reason is simple: the concept was put into use before its content had been agreed upon together.

Borrowed Concepts and Their Limitations

Many organizational development concepts are borrowed. Psychological safety, self-organization, resilience, safer space, servant leadership, shared leadership. Each has its roots in a specific research tradition, in a specific context, in a specific culture. Amy Edmondson developed the concept of psychological safety while studying medical teams. Servant leadership emerged from Robert Greenleaf's reflections in 1970s America. These concepts carry the assumptions of their origins.

Around these concepts, measurement instruments have often been built and "validated" psychometrically. This sounds convincing, but the concept of validation deserves a closer look. The validation samples for many organizational development instruments are based on very narrow groups: American university students, specific age cohorts, participants from a specific cultural background. When an instrument has been validated in such a sample, its psychometric properties hold specifically in that sample. Not automatically in a Finnish expert organization, not in a multicultural team, not among middle-aged managers. More advanced psychometric methods, such as measurement invariance testing, aim to identify and correct this problem, but they remain rare in many areas of organizational development practice.

A broader methodological question has also emerged in psychology research over the past decade. The so-called replication crisis (Open Science Collaboration, 2015) has shown that some classic findings do not replicate cleanly under similar designs, and that some validation methods have been weaker than assumed. This does not mean that all of psychometrics is unreliable, but rather that a "validated instrument" is not an automatic guarantee of quality but a context- and method-dependent matter.

The practical consequence is that the "solid theoretical foundation" of off-the-shelf instruments is genuinely solid only within the limits that the validation sample provides. When the instrument is moved beyond those limits, its results often look precise and reliable while telling very little about what the organization actually wanted to know.

Borrowed concepts and off-the-shelf instruments are still valuable. They offer language, a conceptual framework, and a connection to a research tradition. They can be a good starting point. The important thing is to remember that they are a starting point, not an endpoint, and that their use always requires contextualization to this organization, this culture, and this target group.

Words Build Reality

In the philosophy of language, this is not a new observation. Wittgenstein wrote about "language games": a concept gets its meaning through use, in a specific community, in a specific form of life. The word "safety" is not a neutral container into which the same content is poured everywhere. It gets its meaning in the community that uses it, and in the situations where it is applied.

Berger and Luckmann (1966) presented the basic idea of social constructionism: reality is built in the concepts we use together. When we name something, it begins to exist in shared reality in a new way. When we leave it unnamed, it remains hazy, private, outside the realm of shared meanings.

This also means another thing: if we do not name the phenomena of our own work, someone else does it for us. We then use words that have been formed somewhere else, for some other purpose, from someone else's point of view. The frame of our operating environment is built from these borrowed words, and we begin to see our work through them. This is not a harmless arrangement.

The Culture Behind the Concepts

Borrowed concepts come with two cultural dimensions, both of which are easily overlooked.

The first is the broader cultural context. Concepts shaped in American organizational literature carry the assumptions of American working life: individualism, direct feedback, the importance of hierarchy, performance orientation. A good example is Kim Scott's Radical Candor, which in the American context means open, direct, and caring feedback. In a Finnish work culture, where direct feedback may already feel aggressive by default, the same concept can without contextualization lead either to unnecessary harshness or to a situation where only the "candor" part is taken in without the "radical care" side. Same word, different phenomenon.

The second dimension is organizational culture. Two Finnish companies operating in the same field can have very different ideas about what "open conversation" means. In one, it may be bold challenging and immediate disagreement. In another, it may be considered, subtle, and process-oriented. Both are open conversation in their own context. But if these companies used the same instrument to measure the same concept, they would get incomparable results, even though the numbers might look as if they could be compared.

Organizational culture is itself a language system. It defines which words are used, what they mean, and what can and cannot be said with them. A concept brought in from outside always enters this language system and is shaped by it, whether we are aware of it or not.

What Happens If This Is Not Done?

When a development initiative is launched on a borrowed concept without contextualization, the consequences usually do not appear immediately. They accumulate quietly.

The first consequence is measurement bias. When the instrument is built on a foreign concept, it measures what the original context defined as measurable, not what is meaningful for this particular organization. The results look precise, and they are easy to report to leadership, but they typically describe a different phenomenon than the one the organization thought it was talking about.

The second consequence is action based on wrong conclusions. When biased data is interpreted as truth, the next steps of the development initiative typically target symptoms or even entirely wrong issues. Resources are spent on changes that do not respond to the actual need.

The third consequence is apparent impact. Indicators often show positive development, and the organization reports successes both internally and externally. But if the content of the concept was never agreed upon, "development" happens according to one interpretation, while another part of the organization simultaneously experiences something quite different. Over the longer term, this erodes trust in measurement and in development initiatives more broadly.

These consequences are not speculative. They are common when a development initiative starts quickly and "efficiently" without pausing to consider what the words mean in this organization.

Shared Definition Builds Shared Understanding

A practical conclusion follows. When an organization wants to develop something, it is worth pausing on the words before pausing on the methods. What do we mean when we talk about safer space? How do we recognize it in our own everyday work? What is it not? In which situations is it especially important to us?

These questions may at first reading seem like semantic play, but in fact what is at stake is something quite important: are we using a process in which shared understanding can be built, or are we taking the path of least resistance? Shared understanding is the precondition for all collective action. Without it, each participant operates according to their own internal definition, and the development initiative proceeds seemingly together but in reality in many directions.

Borrowed concepts are a perfectly good starting point in this process. They offer language, a research-based framework, and a point of comparison. But they cannot be the endpoint. The organization must connect the content of the borrowed concept to its own understanding, its own experiences, and the reality of its own operating environment. Only then does the concept begin to mean something shared and worth measuring.

Naming as the First Step

In my previous blog post, I wrote that recognition and naming in practice often function as the first intervention. They direct attention, create shared language, and enable collective examination. This applies to borrowed concepts as well: even when the word is ready-made, its meaning in this context must be built together.

With words we construct reality. Word choices form the frame of our operating environment. If we accept borrowed words as they are, we also accept the frames that come with them. If we name together what the words mean to us, we begin to build our own operating environment, not one borrowed from someone else's work.

What Can You Do Today?

Before your next development initiative, try "conceptual unpacking":

• Ask: What is the origin of this term? (Where was it coined?)

• Ask: What does this mean in our coffee room or in our customer encounters?

• Ask: What is this not? (Define meaning by drawing the boundaries.)

evaluoi.ai is built on this thinking: the concepts used in development initiatives need contextualization, and contextualization needs a dialogical process in which the organization finds its own words for what it actually means.

References

Berger, P. L., & Luckmann, T. (1966). The Social Construction of Reality: A Treatise in the Sociology of Knowledge. Doubleday.

Kenney, M. R. (2001). Mapping Gay L.A.: The Intersection of Place and Politics. Temple University Press.

Open Science Collaboration. (2015). Estimating the reproducibility of psychological science. Science, 349(6251), aac4716.

The Roestone Collective. (2014). Safe space: Towards a reconceptualization. Antipode, 46(5), 1346–1365.

Wittgenstein, L. (1953). Philosophische Untersuchungen [Philosophical Investigations]. Blackwell.