Souvenir craft convergence

Why do the handicrafts sold in tourist towns a thousand miles apart end up resembling each other rather than their own regions?

Walk the craft stalls of a tourist town in the Andes, then in the Balkans, then in Southeast Asia. The motifs differ, but the objects rhyme: small carved animals, painted ceramics of a size that fits a suitcase, textiles cut into scarves and cushion covers, beaded jewellery, magnets, embroidered bags. Regions with entirely unrelated craft traditions have converged on a similar catalogue.

The usual explanation is a story about decline — that authentic traditions have been corrupted, or that everything is now made in the same distant factory. Both of those happen, but neither is required. The convergence would happen anyway, and the interesting question is what produces it even where the objects are genuinely made locally by local artisans.

Start by asking what a traditional craft was originally for. It was made for people who lived there: a cooking vessel of a size that suited local meals, a blanket sized to a local bed, a ritual object whose meaning was legible to neighbours, a tool for a local trade. Its form was answering local questions.

Now change the buyer. The tourist is a different customer with a different set of constraints, and those constraints are almost identical from one country to the next. They must carry the object home in a bag with a weight limit. It must survive being packed. It must be affordable on a holiday budget and cheap enough to buy several as gifts. It must read, at a glance and to someone who knows nothing of the region, as from here — because a souvenir's function is to be recognisable evidence of having been somewhere. And it must be comprehensible without explanation, because there is no neighbour to interpret it.

Those requirements form a filter, and the filter is the same everywhere. Portable, durable, cheap, visually legible as exotic, and self-explanatory. Apply an identical filter to very different starting material and you get outputs that resemble each other more than they resemble their own origins. Nobody has to copy anybody.

Follow the mechanism into the workshop. An artisan with a hundred designs finds that six sell. The six are the ones that passed the filter. Next season they make more of the six and fewer of the rest — not from any loss of integrity, but because they need to eat, and unsold stock is a real cost. Repeat this over some seasons and the surviving repertoire is the filter's output, with everything else pushed to the margins or kept for local use.

Then there is the second, faster loop. Watching what sells is much cheaper than guessing, so traders copy each other's successes within a market, and wholesalers who supply many destinations carry the forms that work in all of them. This accelerates the convergence but it is not its cause — which matters, because it means banning imported stock or licensing local makers does not remove the pressure. The filter is in the buyer, not the supply chain.

One more consequence follows, and it is the counter-intuitive one. Since the object must look regionally distinctive to an outsider, the pressure is not toward blandness but toward a legible, simplified, slightly exaggerated version of a local visual signature. Every region converges on having a strong emblem — and the form that emblem takes converges too, because it must fit in a suitcase. Distinctiveness of pattern, uniformity of object.

"Authenticity" is not the axis this runs on, and treating it as one obscures the mechanism. It is tempting to read convergence as a story of decline from a pure original. But there is no era in which craft was not made for a market and shaped by it — objects made for trade have been adapting to distant buyers for as long as there has been trade. What has changed is the buyer's constraint set, not the fact of having one. The souvenir stall is not craft corrupted; it is craft answering a different question, and answering it competently.

Convergence in form is compatible with real local production. The two get conflated constantly. Imported mass-produced stock certainly exists and certainly increases uniformity. But the argument above needs none of it: independent local artisans facing an identical filter will converge on their own. The test that separates the two is straightforward, and it is worth stating because the policy conclusions differ sharply — if imports were the whole story, restricting them would restore variety, and where that has been tried the effect has been modest.

The strongest evidence is what does not converge. If a shared filter is doing the work, then objects made for local buyers should stay as various as ever, in the very same towns. Ceremonial pieces that never enter the tourist market, work objects, textiles woven to local dimensions for local use — these do not converge, and they sit a few streets from the stalls that do. That is the observation the account has to get right, and it does.

The falsification test. If the mechanism is buyer-side selection, then the convergent catalogue should track the constraints of the buyer rather than the origin of the maker: markets serving visitors who arrive by car with no weight limit should carry larger and heavier objects than markets serving air travellers, from the same craft tradition. If the same restricted catalogue appeared regardless of how buyers travelled and what they could carry, the portability constraint would not be doing the work I have claimed, and the explanation would have to shift toward supply-side copying alone.

• Why the objects that never enter the tourist market stay various, and what that says about the filter.
• How geographical indication schemes try to protect form rather than origin, and where they fail.
• Why the same selection logic shapes airport food and hotel breakfast buffets.