Hard Fork

What Is a Hard Fork?

A hard fork is a change to a blockchain's consensus rules that expands the set of blocks the network considers valid. Because old nodes enforce the original, stricter rules, they reject blocks produced under the new rules. If both old and new nodes continue operating, the result is a permanent chain split: two separate blockchains diverging from a shared history.

The term "fork" comes from software development, where a fork copies a project's codebase and takes it in a new direction. In blockchain, a hard fork copies the entire transaction history up to a specific block height, then the two chains follow separate paths forever. Every holder of the original coin at the fork block automatically holds an equal balance on both chains.

This contrasts with a soft fork, which tightens consensus rules. Soft forks are backward-compatible: old nodes still accept new blocks even without upgrading, so the network remains unified.

How It Works

A hard fork changes the protocol in a way that makes previously invalid blocks valid. The mechanics unfold in a predictable sequence:

1. Developers propose a rule change that loosens existing consensus constraints (for example, increasing the maximum block size from 1 MB to 8 MB)
2. A new software version implementing the change is released with a scheduled activation at a specific block height
3. At the activation block, miners running the new software produce a block that is valid under the new rules but invalid under the old rules
4. Upgraded nodes accept this block and continue building on it. Old nodes reject it and continue building on the last block they considered valid
5. The network splits into two chains, each with its own set of miners, nodes, and transaction history from the fork point onward

Hard Fork vs. Soft Fork

The fundamental distinction is about the direction of rule changes:

Replay Protection

When a hard fork creates two chains, a transaction valid on one chain may also be valid on the other. This is called a replay attack: someone could broadcast your transaction on the chain you did not intend. Well-designed hard forks include replay protection, which modifies the transaction format so that transactions are only valid on one chain. Without it, users must use "coin splitter" tools to safely separate their holdings.

Bitcoin's Hard Fork History

Bitcoin has never undergone a deliberate, community-approved hard fork on its main chain. The forks that created alternative Bitcoin networks were all contentious: minority factions who disagreed with the development direction chose to split away.

The Accidental Fork of 2013

On March 12, 2013, Bitcoin experienced an unintentional chain split. Bitcoin Core version 0.8.0 had switched its database from Berkeley DB to LevelDB, but a mined block contained more transaction inputs than Berkeley DB could handle. Nodes running v0.8 accepted the block while v0.7 nodes rejected it, splitting the network.

Major mining pools voluntarily downgraded to v0.7, reconverging on the shorter chain. A double spend did occur during the split window. The incident, documented as BIP 50, became an early lesson in the dangers of accidental consensus divergence.

Bitcoin Cash (August 2017)

Bitcoin Cash (BCH) forked at block 478,558 on August 1, 2017. A faction of developers, miners, and businesses led by figures including Bitmain's Jihan Wu wanted to increase the block size from 1 MB to 8 MB (later 32 MB) to handle more transactions on-chain, rather than relying on the SegWit approach. BCH removed SegWit and increased the block size limit directly.

Bitcoin Gold (October 2017)

Bitcoin Gold (BTG) forked at block 491,407 on October 24, 2017. Its goal was to re-decentralize mining by replacing Bitcoin's SHA-256 proof-of-work algorithm (which favors ASIC miners) with Equihash, which was designed to be GPU-friendly.

Bitcoin SV (November 2018)

Bitcoin SV (Satoshi's Vision) forked from Bitcoin Cash at block 556,767 on November 15, 2018. Led by Craig Wright and Calvin Ayre, the fork argued that BCH's scaling was still insufficient and pushed for a 128 MB block size limit (later effectively unlimited). BSV represented a fork of a fork: it split from BCH, not directly from Bitcoin.

The Blocksize War

The Blocksize War (2015 to 2017) was the defining political conflict in Bitcoin's history. It determined whether Bitcoin would scale through on-chain block size increases (requiring a hard fork) or through off-chain solutions enabled by soft forks. The outcome shaped Bitcoin's governance philosophy permanently.

The conflict centered on a simple technical question: should the 1 MB block size limit, introduced by Satoshi Nakamoto in 2010, be raised? One camp argued that larger blocks would allow more transactions per second, keeping fees low. The other camp argued that larger blocks would centralize the network by increasing the cost of running a full node, and that scaling should happen on Layer 2 protocols instead.

Key events included the Hong Kong Roundtable (February 2016), where miners and developers failed to agree on a path forward, and the New York Agreement (May 2017), where over 50 companies signed onto SegWit2x: a plan to activate SegWit followed by a hard fork to 2 MB blocks. The hard fork component was ultimately canceled in November 2017 due to lack of consensus.

The conflict was resolved by user-activated soft fork (UASF) pressure. Individual node operators signaled they would reject blocks that didn't support SegWit, demonstrating that network governance ultimately rests with node operators, not miners or corporations. SegWit activated on August 24, 2017 at block 481,824 as a soft fork, and the Blocksize War ended.

Why Bitcoin Avoids Hard Forks

The Blocksize War established a strong cultural norm against hard forks in Bitcoin. Several principles drive this resistance:

• Network fragmentation: hard forks can permanently split the community, hashrate, and economic network. Bitcoin Cash demonstrated that even forks with significant corporate backing end up with a fraction of the original network's value.
• Mandatory upgrades: hard forks require every node operator to upgrade or be ejected. With tens of thousands of independently operated nodes worldwide, mandatory universal upgrades conflict with Bitcoin's decentralization ethos.
• Precedent risk: each hard fork lowers the barrier for the next one. The Bitcoin community views resistance to protocol changes as a feature of sound money, not a bug.
• Soft forks suffice: upgrades like SegWit and Taproot demonstrate that meaningful protocol improvements can be deployed without splitting the network.

This philosophy extends to current development. Active proposals like OP_CTV, OP_CSFS, and OP_CAT are all designed as soft forks: backward-compatible changes that add new functionality through covenants without risking a chain split.

How Soft Forks Achieve Similar Goals

The SegWit upgrade illustrates how soft forks can accomplish what might otherwise require a hard fork. The goal was to increase Bitcoin's transaction throughput, which the block size increase camp proposed solving via a hard fork.

Instead, SegWit restructured transactions by moving witness data (signatures) into a separate structure. It replaced the 1 MB block size limit with a 4 million weight unit system, where witness data weighs less than transaction data. This effectively allows blocks of 2 to 4 MB while remaining backward-compatible: old nodes see valid blocks under 1 MB because they strip out witness data they don't understand.

Beyond throughput, SegWit fixed transaction malleability, enabling the payment channel designs that underpin the Lightning Network. It also introduced script versioning, making future soft fork upgrades easier to deploy.

Hard Forks in Other Ecosystems

Not all blockchain communities share Bitcoin's aversion to hard forks. Ethereum uses planned hard forks as its standard upgrade mechanism, deploying major protocol changes like Constantinople (2019), London (2021), and The Merge (2022) through coordinated hard forks with near-universal node adoption.

The most notable contentious hard fork outside Bitcoin was Ethereum's DAO fork on July 20, 2016 at block 1,920,000. After an attacker exploited a smart contract vulnerability to drain roughly 3.6 million ETH (about $60 million at the time), the Ethereum community voted to hard fork and reverse the theft. Roughly 10 to 15 percent of the community rejected this on immutability grounds, continuing the original chain as Ethereum Classic (ETC).

This contrast highlights a philosophical divide: Ethereum treats hard forks as routine maintenance, while Bitcoin treats them as existential threats. Both approaches have tradeoffs: Ethereum iterates faster but requires more trust in the upgrade coordination process, while Bitcoin evolves slowly but maintains stronger guarantees about rule stability.

Why It Matters

Understanding hard forks is essential for anyone building on or holding Bitcoin. Hard forks affect how you secure your assets during chain splits, how you evaluate new networks claiming the Bitcoin name, and how you understand Bitcoin's governance model.

For developers, the Bitcoin community's preference for soft forks shapes the design space for new features. Proposals that require hard forks face an extremely high bar, which is why innovations like Layer 2 solutions and covenant proposals focus on building within Bitcoin's existing consensus rules or extending them through backward-compatible soft forks. Protocols like Spark take this approach, delivering new capabilities on top of Bitcoin without requiring changes to the base layer's consensus rules.

Risks and Considerations

For Holders

During a hard fork, holders of the original coin receive equivalent balances on both chains. However, accessing forked coins requires careful handling. Without replay protection, spending on one chain could unintentionally spend on the other. Users should wait for clear guidance, use coin-splitting tools, and avoid transacting during the immediate fork period.

For Miners

Miners must choose which chain to dedicate hashrate to after a fork. If a minority chain retains the same proof-of-work algorithm, it may experience extremely slow block times until its next difficulty adjustment, leaving transactions unconfirmed for extended periods.

For the Network

Contentious hard forks fragment developer talent, community attention, and economic value. Bitcoin Cash, Bitcoin SV, and Bitcoin Gold collectively represent billions of dollars in forked value, but none has approached Bitcoin's network effects, hashrate, or adoption. Each fork creates confusion about which chain is "the real Bitcoin" and dilutes the brand among new users.

Security During Transition

Immediately after a hard fork, both chains may have reduced hashrate, making them more vulnerable to double-spend attacks. The minority chain is especially vulnerable if it retains the same proof-of-work algorithm, as miners can switch between chains. Bitcoin Gold suffered multiple 51% attacks in 2018 and 2020, resulting in millions of dollars in double spends.

This glossary entry is for informational purposes only and does not constitute financial or investment advice. Always do your own research before using any protocol or technology.