The concept is developed through a wide range of examples — the lines of writing in medieval manuscripts, the lines of walking in indigenous navigation, the lines of drawing in artistic practice, the lines of song in oral traditions. In every case, Ingold finds evidence for the same typological distinction: some lines connect, others grow. And he finds evidence for a broader historical shift: across many cultures, lines have been progressively transformed from lines of making into lines of transport, as modern life has privileged arrival over journeying.
The AI moment accelerates this shift dramatically. The prompt-execute cycle is, in Ingold's terms, the purest possible line of transport: origin (prompt) and destination (output) are the only relevant points, and the space between them is traversed instantly and invisibly. The user does not walk through the problem space; she is delivered across it. The product is comparable to what a walking practitioner might have produced, but the line has a different structure, and the difference has consequences for what the maker learns.
The specific loss is the loss of surprise. Lines of making generate discoveries that could not have been anticipated from the starting point. The wayfarer gets lost, finds unexpected openings, doubles back, encounters detours that turn out to be the real path. This capacity for genuine surprise is what makes lines of making generative of genuinely new work. Lines of transport cannot produce this kind of surprise because the route has been calculated. The user of AI may be surprised by what the tool produces, but the surprise is surprise at the tool, not surprise within the creative process.
The most theoretically interesting aspect of the line of making concept is that it dissolves the image of the creator as a sovereign planner. The wayfarer does not plan her path; she grows it. This reframes authorship. The author of a line of making is not the originator of a pre-existing plan; she is the walker of a path that emerged through the walking. This reframing is uncomfortable for the conventional picture of the autonomous creator, and it resonates with the author's honest reports, in You On AI, that the best work emerged from collaboration processes whose outcomes neither he nor Claude had predicted.
The typology of lines is developed most fully in Lines: A Brief History (2007), though the distinction between wayfaring and transport appears in Ingold's earlier work. The Deleuze-Guattari concept of 'lines of flight' (lignes de fuite) is acknowledged as a source, though Ingold's usage is distinct: for Deleuze and Guattari, a line of flight is an escape from a striated space, whereas for Ingold, a line of making is a generative movement through a landscape.
The ethnographic evidence comes from a wide range of sources, including Ingold's own fieldwork, anthropological accounts of indigenous navigation, and historical studies of lines in medieval manuscript culture.
Two kinds of lines. Lines of transport connect predetermined points; lines of making grow through a landscape without a fixed destination.
The landscape teaches. The wayfarer's attention to terrain features produces both the path and the perceptual capacities the path leaves behind as residue.
Surprise is the signature of making. Lines of making generate genuinely unpredictable outcomes; lines of transport cannot, because the route is calculated.
Authorship is walking, not planning. The maker of a line of making is the walker of a path that emerged through the walking, not the originator of a pre-existing plan.
AI is structurally biased toward transport. The prompt-execute cycle is the purest form of transport line; correspondence-like collaboration is possible but requires deliberate resistance to the medium's dominant mode.