Lightning materially improves payment privacy

Payments routed over Lightning are materially more private than on-chain transactions, because intermediate nodes do not learn who the sender or the recipient is.

For (strongest version as argued): The routing protocol is built for this. Communication between nodes is encrypted point-to-point, and the sender constructs the route as nested encrypted instructions — an onion scheme (Sphinx) in which each hop can decrypt only its own portion. An intermediary learns the previous and next hop and nothing else: not the other nodes, not the length of the path, not its own position in it. Paths are fixed at 20 hops and padded with random data precisely so that position cannot be inferred — to every node but the final recipient it appears there are always 20 more hops to go. A network-level observer cannot associate packets from different parts of the path. Unlike Tor, there are no exit nodes to place under surveillance. The consequence the notes draw is fungibility: surveillance and blacklisting become much harder to apply.

Against (strongest version as argued): The comparison is asserted, not measured, and the same source supplies the counterweights. A channel does not exist until a funding transaction is transmitted and mined, and it ends with either a cooperative settlement or a unilateral commitment transaction — so channel endpoints and capacities are anchored on-chain and publicly visible at both ends of a channel's life. The privacy is also asymmetric in a way the summary hides: the full path is known to the sender's node. Whatever protection exists runs against intermediaries, not against the payer. And the properties claimed are design intentions of a protocol description; whether routing nodes can in practice correlate payments by timing and amount, and whether a public channel graph permits balance probing, are empirical questions the notes do not address at all.

Nuance: Three different privacy questions are routinely merged. (1) Privacy from intermediate routing nodes — the source's actual claim, and its strongest. (2) Privacy from chain surveillance — only partial, since opens and closes are on-chain by construction. (3) Privacy from the counterparty or the payer — weakest, since the sender builds the route. A speaker who establishes (1) has not established (2) or (3). Note also that these notes describe an early design of the protocol and do not state an edition or date; treat the mechanism description as a snapshot of the book's writing rather than a statement about the network as it runs today. This entry rests on one owner-research file with no external corroboration and has not been cross-checked against a primary source — it caps at CONTESTED for that reason (corpus/INGEST_PLAYBOOK.md).

Common misstatements: "Lightning payments are off-chain, so they leave no trace" (opening and closing a channel are on-chain transactions — the source's own funding-transaction description says the funding transaction is the one transaction that must be mined). "Lightning is anonymous" (the claim on offer is privacy from intermediaries, which is not anonymity). "It's Tor without the exit-node problem, so the deanonymization attacks don't apply" (the source does draw the exit-node contrast; it does not follow that every correlation attack is thereby ruled out).