Bitcoin’s consensus mechanism works because, most of the time, everyone agrees on the rules. Nodes collect transactions in the mempool, miners compile blocks, submit them, nodes verify them, the chain extends, and miners get a reward for their efforts and energy.
Everyone is happy!
When a proposal like BIP 110 — the “Reduced Data Temporary Softfork” aimed at restricting arbitrary data storage — moves toward activation, that agreement gets tested directly. Nodes start running different software, miners are asked to signal which ruleset they’re mining under, and for a period of time, Bitcoin’s single canonical chain is genuinely at risk of splitting into two.
If this is your first contentious soft fork, you might not be familiar with the rules of the game, how that split works, which side tends to win, and why your own transactions could get caught in the crossfire.
How a Soft Fork Actually Diverges the Chain
A soft fork attempts to change Bitcoin’s consensus rules — it makes previously valid transactions under the old rules or blocks invalid, without changing anything about what old, non-upgraded nodes will accept.
That asymmetry is the whole point of a soft fork: blocks that follow the new rules are still valid under the old rules too, so non-upgraded nodes keep accepting them without needing to update.
In theory, this makes soft forks “backwards compatible” and lower-risk than a hard fork.
In practice, activation still requires the network to reach a coordinated flip.
With BIP 110 specifically, the mechanism under discussion is a User-Activated Soft Fork (UASF) — a flag-day approach where nodes running BIP 110-compliant software begin enforcing the new rules at a predetermined block height, regardless of whether miners have signalled support.
After the flag date, miners are then asked to signal their intent by including version bits or specific markers in the blocks they mine, effectively declaring which ruleset they’re building on top of.
The divergence happens the moment a miner producing blocks under the old rules creates a block that BIP 110-enforcing nodes reject — for example, a transaction using the wider OP_RETURN or larger witness data BIP 110 restricts.
- Nodes running the new rules refuse to build on top of that block.
- Nodes still running old rules accept it happily.
From that block onward, two separate chains exist, each considered valid by a different subset of the network, sharing the same history up to the fork point but diverging afterwards.
Which Direction the Fork Goes, and Why?
Bitcoin’s consensus rule is simple.
The chain with the most accumulated proof-of-work, as recognised by the nodes and economic activity that matter to you, wins. This is where things get genuinely uncertain during a UASF, because the two forces that determine outcome — hashrate and economic weight — don’t have to point the same direction.
If a strong supermajority of miners signal for BIP 110, the new-rules chain accumulates more proof-of-work fastest, and it becomes the chain that non-upgraded nodes and lagging exchanges eventually recognise as the “real” Bitcoin, since it’s simply the longest valid chain by the time anyone checks.
But because BIP 110’s proposed activation method deliberately doesn’t wait for a miner supermajority the way traditional soft forks like SegWit historically leaned on, it’s entirely possible for a chain enforcing the new rules to exist with a minority of hashrate behind it, at least temporarily.
In that scenario, the new-rules chain can still “win” in the sense that matters most — recognised by exchanges, wallets, and economic activity — provided enough of that economic majority runs upgraded nodes and simply refuses to transact on, or accept payment from, the old-rules chain.
This is the dynamic that played out with SegWit’s 2017 UASF: the threat of economic nodes rejecting non-compliant blocks was enough to bring miner signalling in line before the flag day even arrived, avoiding a prolonged split.
There’s no guarantee BIP 110 resolves the same way.
What Happens To The Shorter Chain During The Chain Split?
The blocks on the shorter chain are permanently discarded by the network through a process called a chain reorganisation (reorg), causing all unique blocks on that shorter chain to become stale blocks (often called orphaned blocks).
Mechanics of the Chain Selection
Bitcoin nodes always follow the longest chain rule, which technically means the chain with the most accumulated proof-of-work.
[Block 101] -> [Block 102] (Shorter Chain - DISCARDED)
/
[Block 100]
\
[Block 101] -> [Block 102] -> [Block 103] (Longer Chain - ACCEPTED)
- Network alignment: Nodes automatically switch to the longer chain.
- Block erasure: Shorter chain blocks lose validity completely.
- Mining rewards: Coinbase rewards on the shorter chain vanish.
Transactions inside the discarded blocks face two possible paths:
- Already included: Most transactions are already in the longer chain. These remain completely unaffected.
- Returned to mempool: Unique transactions go back to the pool. Miners can then add them into future blocks.
- Double spends: Transactions conflicting with the longer chain become invalid
What Happens to the Losing Fork if it Hard Forks?
If the split doesn’t quickly reconcile, one chain typically becomes the dominant one — carrying the ticker “BTC,” the liquidity, the exchange listings, and the market’s attention — while the other survives, if at all, as a minority chain with its own name, its own reduced hashrate, and dramatically less economic activity.
History offers a clear precedent for a chain split or hard fork in Bitcoin Cash. The chain split from Bitcoin in 2017 over a block size disagreement and continues to exist, but at a tiny fraction of Bitcoin’s price, hashrate, and relevance.
The losing fork doesn’t disappear technically — the blockchain and the coins on it still exist, and anyone holding pre-fork Bitcoin technically holds a balance on both resulting chains — but it can become illiquid, thinly traded, and vulnerable to further reorganisation or abandonment as miners stop finding it profitable to secure.
The Implications of Backing the Wrong Fork
If you transact, hold, or run infrastructure on the fork that ultimately loses economic dominance, the practical consequences are severe. Any Bitcoin balance you have is still there, but its market value can collapse relative to the dominant chain almost immediately once the split is recognised as permanent.
Merchants and exchanges that continued accepting payments on the losing chain during the window of confusion may find those transactions worthless once the network settles on the winning fork.
Businesses running nodes or wallet infrastructure that don’t upgrade in time can find themselves silently processing transactions on a chain the rest of the world has already abandoned — accepting payment for goods or services in a currency that, from the market’s perspective, no longer functions as Bitcoin.
Why Soft Forks Are Risky for Layer 2s Like Lightning
Lightning Network channels depend on pre-signed transactions — commitment transactions and penalty transactions that both parties hold in reserve, ready to broadcast if the other party cheats or the channel needs to close.
During a contested soft fork, if the rules governing which transactions are valid change out from under an open channel, there’s real risk that a pre-signed transaction one party is counting on to enforce a channel closure becomes invalid on the chain that ultimately wins, or behaves unexpectedly on a chain it wasn’t built for.
Lightning Node operators and routing nodes also have to decide, in real time, which chain to keep their Lightning infrastructure pointed at, and a wrong guess can mean force-closing channels, disrupted routing, or funds temporarily stuck while the dust settles.
This is why competent Lightning nodes tend to pause channel opens and treat force-closes cautiously around contentious fork events, rather than operating normally through the uncertainty.
Why You Should Be Cautious Transacting During a Fork
The core advice during any live, contested fork event is simple: slow down.
Wait for additional confirmations before considering a transaction final, avoid moving large amounts until the dominant chain is clearly established, and be sceptical of any payment you receive until you’re confident which chain — and whose nodes — actually recognise it.
Exchanges often pause deposits and withdrawals during exactly these windows for the same reason.
A transaction that looks perfectly confirmed on one chain can simply not exist on the other, and if you can’t yet tell which chain is going to matter, you can’t yet tell whether what you’re holding is real.
Patience, not speed, is what protects you when Bitcoin’s rules are being actively contested.
There Can Only Be One
Temporary consensus fractures will happen from time to time; it’s baked into the pie of programmatic money. Software will need to upgrade and improve, or become fragile or at risk, even though in most cases ossification is the conservative and safe protocol.
Bitcoin’s incentive structure ultimately has a plan for upgrades and forces a resolution.
A split network creates friction, split liquidity, and heightened risks for users, miners, and infrastructure providers alike. Because market value, hash rate, and network effects naturally gravitate toward the chain backed by the economic majority, minority forks rarely maintain parity for long.
Whether resolved through miner compliance, economic alignment, or a permanent divergence into a secondary altcoin, Bitcoin’s ultimate strength lies in its ability to enforce a single, unforgeable ledger—proving that, in the end, there can only ever be one canonical chain.
