The very first public response to Satoshi's announcement of Bitcoin was an expression of doubt about scalability:
We very, very much need such a system, but the way I understand your proposal, it does not seem to scale to the required size.
...
To detect and reject a double spending event in a timely manner, one must have most past transactions of the coins in the transaction, which, naively implemented, requires each peer to have most past transactions, or most past transactions that occurred recently. If hundreds of millions of people are doing transactions, that is a lot of bandwidth - each must know all, or a substantial part thereof.
https://www.mail-archive.com/cryptography@metzdowd.com/msg09...
And the comment was spot-on. Naively implemented, Bitcoin can only scale linearly by putting every transaction onto the block chain. Double the transaction capacity means doubling the size of blocks.
Lightning Network scales the Bitcoin network by offering a secure method for keeping many, if not most, transactions off the block chain. This is done through a clever use of Script, Bitcoin's built-in programming language.
Two parties who want to transact at high speed jointly lock up some money with an on-chain transaction (thereby creating a "payment channel"). Then they send each other half-signed transactions spending the locked funds. Each transaction represents a kind of private ledger. Although any of these transactions can be published at any time, it will usually be mutually beneficial to avoid doing so and continue passing half-signed transactions back and forth. When it's time to call it quits, the last transaction is published.
In other words, the intermediate transactions never need to hit the block chain. Only the final settlement transaction needs to be published.
Lightning takes this one step further by providing a mechanism whereby a party can pay another party on behalf of a third party. This means that you can make Lightning payments without necessarily having an open channel with your intended payee. You just need an open channel and the network routes your payment.
Lightning Network has been in development for years. The idea has been repeatedly disparaged as "vaporware." The first mainnet release is a big deal because that characterization no longer applies. It also means we're going to see a real-world test of an idea that up until now hasn't been widely-tested.
The Bitcoin Cash split was due in large part to a group of Bitcoiners rejecting Lightning as a scaling option. They tend to believe not only that Bitcoin can scale by increasing block size, but that this is the best way.