MIVORA Start learning free

Path Dependence: Why History Sticks

Once a system settles on a way of doing things, it can get locked in — so the future is shaped not just by what’s best, but by what came first.

The Future of Technology · Lesson 36 · 10 min read

Last lesson: once a system tips into a new state, it’s often hard to reverse. This lesson explains a profound consequence — the future is shaped by history in a way that isn’t always “the best option wins.” Sometimes an early, even arbitrary choice gets locked in and constrains everything that follows, long after better alternatives appear. Understanding this stops you from assuming technology and society always converge on the optimal answer — they often converge on the answer that got there first and stuck. Hold the question: why do we still use the “QWERTY” keyboard layout, designed for typewriters over a century ago, even though it wasn’t designed for speed?

Path dependence: how you got here constrains where you can go

Path dependence means that the history of how a system developed — the specific path it took — constrains its present and future, sometimes more than the merits of the current options do. Once a choice is made and everyone builds around it, that choice gets locked in: it becomes the standard, and switching away — even to something better — becomes costly and coordinated enough that it usually doesn’t happen. QWERTY is the classic example: a layout chosen for old mechanical reasons, kept not because it’s best but because everyone learned it, every keyboard uses it, and switching everyone at once is nearly impossible. History, not optimality, is holding it in place. “Where a system can go next depends on where it’s been” is the essence of path dependence.

Why lock-in happens: switching costs and coordination

Lock-in comes from two forces you’ve met. Switching costs: once you (and everyone) have invested in a standard — learned it, built tools around it, made everything compatible — moving to an alternative means throwing away that investment, which is painful enough that people don’t. Coordination (network effects, lesson 33): a standard is valuable because everyone else uses it, so switching only pays if everyone switches together — and getting everyone to move at once is a brutal coordination problem (next lesson). Together these create a powerful status-quo trap: even when a clearly better option exists, no individual can afford to switch alone, and no one can make everyone switch together, so the incumbent persists. This is the flip side of the tipping-point flywheel (lesson 35): the same feedback that made a standard dominant now makes it sticky.

Worked example
Why a better standard often loses:
• A new keyboard layout is provably faster. Should everyone switch? Individually: no — you’d be slower for weeks, and every keyboard, tutorial, and habit is QWERTY.
• Collectively it might be worth it, but only if everyone moves at once — which no one can coordinate.
• So the better option loses to the locked-in one. The winner was decided by history and coordination, not merit.

What this means for reading the future

Path dependence carries three lessons for thinking ahead. First, don’t assume the best technology wins — the first-to-lock-in often does, so when predicting a “standards war,” ask who’s reaching critical mass, not just who’s best (this is why the early moments of a new technology matter so disproportionately — that’s when the path gets set). Second, some things are stuck, and that’s okay to notice — a lot of “why do we still do it this dumb way?” is path dependence, not stupidity; the switching cost genuinely exceeds the benefit. Third, and hopefully: lock-in can break at rare moments — a big enough discontinuity (a new platform, a generational shift, a technology so much better it overcomes the switching cost) can reset the path, which is exactly when long-frozen things suddenly change. The honest forecaster holds both truths: history has real inertia (don’t expect the optimal to just win), and inertia occasionally breaks (watch for the moments that reset the path). Where a system has been genuinely shapes where it can go — but not forever.

An everyday analogy

Imagine a river carving a valley. The very first trickle of water follows some small, almost random dip in the land — but as it flows, it deepens that channel, and now more water follows it, deepening it further, until a great river runs there. Centuries later, the river is locked into a path that a tiny early accident chose, and it would take an earthquake to move it, even if a straighter route exists right alongside. Technologies and institutions carve their valleys the same way: an early, sometimes arbitrary choice gets deepened by everyone flowing through it until it’s a canyon no individual can climb out of — and only a rare upheaval reroutes the river.

Worked example
Spotting path dependence and its limits:
1. Is a standard dominant because it’s best, or because it got there first and everyone built around it? (Often the latter.)
2. What locks it in: switching costs (wasted investment) + coordination (need everyone to move together)?
3. For forecasting a standards war: watch who reaches critical mass early, since the path may set then — not who’s technically best.
4. Ask what could break the lock-in: a big discontinuity (new platform, generational shift, a vastly better option). Inertia is real but not eternal.

This is the reading. The interactive version — active-recall quiz, a hands-on experiment you run in your own AI, and an earned mastery check — is free in the app.

Start this lesson free →