Klaus Krippendorff has been a foundational influence on my understanding of design since my MPhil. His work sits at the intersection of design, cybernetics, and communication theory - a combination that becomes increasingly relevant as design moves beyond physical products into information systems, services, and social interventions.
I've been returning to his writing as I try to make sense of my current situation at SCÖ. The project I've joined expects me to explore "data science" and "federated learning" for vocational rehabilitation. But the more time I spend here, the more I realise we're not designing products or even interfaces - we're working at the level of discourse, trying to shape how different stakeholders understand what "data science" could mean in this context.
Krippendorff's "trajectory of artificiality" offers a framework for understanding this shift, and why it requires fundamentally different design competencies.
The Trajectory
In his 2011 paper "Principles of Design and a Trajectory of Artificiality", Krippendorff maps how design problems have evolved through five major classes, each building upon and rearticulating the preceding:
1. Products - largely industrial and material, designed for utility, functionality, and universal aesthetics. The designer's responsibility coincides with manufacturing and terminates with the end-product. The classic principle: form follows function.
2. Goods, services, and identities - market and sales-driven. Utility becomes secondary to meaning, recognition, attraction, and consumption. These "reside largely in the memories, conceptions, attitudes, preferences, and loyalties of large populations of people" (Krippendorff, 2011). Design becomes concerned with symbolic qualities shared within targeted consumer groups.
3. Interfaces - computers, simulators, and control devices as extensions of human conceptions. Miniaturisation and digitisation made internal functionality "nearly incomprehensible to ordinary users, if not irrelevant", shifting attention from architecture to interactive languages. Criteria become: human-machine interactivity, understandability, reconfigurability, adaptability.
4. Multi-user systems (networks) - facilitating coordination across space and time. Information systems, communication networks, the Internet. Designers become concerned with informaticity, connectivity, and mutual accessibility. Projects emerge around technologies but are "embodied in human communicative practices".
5. Discourses - living in communities of people who collaborate in their production. The design of discourses focuses on "their generativity (their capacity to bring forth novel practices), their rearticulability (their ability to provide understanding), and on the solidarity they create within a community" (Krippendorff, 2011).
What strikes me about this trajectory is that it's not simply a historical progression - though there is a historical dimension - but rather an increasing complexity of the materials designers work with. At each level, the "material" becomes less tangible and more social.
From Mono-Logic to Multi-Logic
Krippendorff draws a crucial distinction between first-order and second-order understanding.
First-order understanding is the kind engineers need: understanding how things work according to specifications. It assumes everyone sees artefacts the same way. It's "mono-logic" - a single rationality that everyone should comprehend and enact.
Second-order understanding is understanding of understanding. It assumes that others' understanding is potentially different from one's own. It's "multi-logic" - recognising the multiple rationalities that different stakeholders bring to bear on artefacts.
As Krippendorff puts it in The Semantic Turn:
"We must realize that the understanding of someone else's understanding of something is qualitatively different from the understanding of that something. Understanding someone else's understanding is an understanding of understanding, an understanding that recursively embeds another person's understanding in one's own, even if, and particularly when, these understandings disagree, contradict one another, or are thought by one to be wrong or appallingly unethical" (Krippendorff, 2005, p. 17).
This distinction matters enormously for design practice. At the product level, first-order understanding might suffice - you can design a functional object based on engineering rationality. But as you move up the trajectory toward interfaces, networks, and discourses, second-order understanding becomes essential. You cannot design an interface without understanding how users will understand it. You cannot design a network without understanding how different stakeholders will use it in their own terms. You cannot participate in a discourse without understanding how your contributions will be understood by others.
Meaning as the Only Reality that Matters
One of Krippendorff's fundamental insights, developed through his work on product semantics, is captured in what he calls an "irrefutable axiom of design":
"No artifact can survive within a culture without being meaningful to those that can move it through its defining process" (Krippendorff, 2011).
This inverts the classic design principle. Instead of form follows function, the principle becomes form should follow meaning. Design has to make sense to others.
This represents a recognition that artefacts exist within webs of meaning that are socially constituted. A knife means something different when a friend asks you to cut bread than when a stranger holds one while demanding money. The object is the same; the meaning - and therefore the reality for the people involved - is entirely different.
For designers, this means acknowledging that "the diversity of individual (user) conceptions matter as much as if not more than the (techno-)logic of designers and engineers" (Krippendorff, 2011). The rationality embedded in an artefact by its designers may not be the rationality through which users engage with it.
Design Must Be Delegated
Another principle that follows from the trajectory: as design problems become more complex, designers lose their monopoly on design decisions.
"In the information age, designers can no longer claim a monopoly on design decisions. In fact, users often are better designers of their own worlds than professionals, and even if they are not they tend to take pleasure in doing so with their own criteria" (Krippendorff, 2011).
Human-centered design, in this view, means "leaving artefacts underspecified, open, enabling interested others to be designers of their own worlds, in their own terms, and towards satisfying their own needs".
This challenges the traditional conception of design expertise. Krippendorff notes that desktop publishing made graphic designers its first victims - computational technology enabled ordinary users to do what professionals had claimed as their domain. Design cannot protect itself through licensing the way medicine does. "Designing is a fundamentally human activity in everyday life".
The implication: professional designers can only stay ahead by pursuing the trajectory, working on problems that others haven't yet engaged with. "Design amounts to boldly walk where others have not dared to tread".
Heterarchy, Not Hierarchy
Herbert Simon's influential work on the architecture of artificial systems celebrated hierarchy and mono-logical rationality. Simon imagined systems designed top-down, with a single rationality governing their logic.
Krippendorff argues this no longer holds:
"By their very nature, information networks must afford the considerable conceptual diversity of its many users, enable groups to realize themselves in them, and allow individuals to read data as information in their own terms. The success of the good old telephone network and now the Internet lies precisely in the fact of no restrictions on what can be said" (Krippendorff, 2011).
Design needs to operate with heterarchical conceptions, embrace diversity of meanings, and negotiate outcomes with many interested parties. "Chaos, heterarchy, diversity, and dialogue are the new virtues that design must embrace today".
This connects to the stakeholder perspective: "Artefacts (are) create(d in) networks of stakeholders". Before a design comes to fruition, it moves through different lives - as ideas, sketches, prototypes, production drawings, sales franchises, packaged goods, status symbols. At each stage, different people claim different stakes. Designers are necessarily part of these stakeholder networks.
Exploration, Not Re-Search
Krippendorff makes a provocative distinction between exploration and research. Design creates new futures by intervening in the present. Scientific research "favors history and thrives on observations of constraints, what cannot or has not been done".
The hyphen in "re-search" emphasises the etymology: a repeated search for explanations of given data, their careful re-examination. Science assumes the world exists as observed and that explanatory logic governs the future as demonstrated in the past.
But along the trajectory of artificiality, "nothing ever repeats itself". Designers who rely too heavily on data-driven research "systematically and methodically fossilize creative deviations from observed history".
Instead:
"Designers have to explore the present for what is variable, combinable into new artefacts, fusible into new technologies in order to reach desirable futures for targeted communities. 'Scouting', 'wayfinding', or 'trail-blazing' are other words for how designers need to empirically explore the present to inform themselves about what can be done and how" (Krippendorff, 2011).
This resonates with Donald Schön's conception of the reflective practitioner - a paradigm of inquiry that acknowledges the dynamics that design activity sets in motion.
Artefacts Survive or Fail in Language
Perhaps Krippendorff's most radical insight:
"To succeed, artifacts must survive in language or fail" (Krippendorff, 2011).
Contemporary artefacts have become "language-like" - recombined into numerous forms, acquiring new meanings in diverse contexts of use, variously rearticulable by different users. The environmental complexity we face is the complexity of our language.
Design specifications are formulated in language. Marketing, advertising, and user practices occur in language. "Virtually all design is accompanied by talk". We know of others' perceptions, feelings, acceptance and rejection only through what they tell us.
This doesn't downplay visual perception, tactile experience, emotion, or kinesthetic sense. But it does mean that design discourse - the constrained conversation through which a design community accomplishes its work - is what keeps designers together, generates opportunities, and enables the education of future practitioners.
What This Means for SCÖ
Reading Krippendorff in my current context clarifies something important.
The ADAPT project isn't asking me to design a product. It's not even asking me to design an interface or a network in the technical sense. It's asking me to participate in a discourse - to help shape how "data science" and "federated learning" are understood within a network of stakeholders with very different conceptions of what these terms mean.
At the discourse level, the design "materials" are narratives, meanings, understandings. The design work is about generativity (what new practices could these concepts enable?), rearticulability (how can they be understood across different stakeholder groups?), and solidarity (what shared commitments could they create?).
This is why traditional design methods feel inadequate here. Concept maps, journey maps, and service blueprints are tools developed largely for the interface and network levels of the trajectory. They help visualise interactions, identify touchpoints, and coordinate stakeholders around shared representations.
But discourse-level design requires something different - second-order understanding of how different stakeholders understand the concepts at play, and the capacity to participate in reshaping those understandings. It's less about making artefacts and more about making meaning.
Whether I can actually do this - whether anyone can, in this context - remains to be seen. But at least I now have a framework for understanding what kind of problem I'm facing.
References
Krippendorff, K. (1997). A trajectory of artificiality and new principles of design for the information age. In K. Krippendorff (Ed.), Design in the Age of Information: A Report to the National Science Foundation. University of Pennsylvania.
Krippendorff, K. (2005). The Semantic Turn: A New Foundation for Design. Taylor & Francis/CRC Press.
Krippendorff, K. (2011). Principles of Design and a Trajectory of Artificiality. Journal of Product Innovation Management, 28, 411-418.
Schön, D. A. (1983). The Reflective Practitioner: How Professionals Think in Action. Basic Books.
Simon, H. A. (1969). The Sciences of the Artificial. MIT Press.