Oh, the irony...
Nice article, Richard, thank you... though my name is Wilton with a "t" (not to be confused with Steve, who is Wilson with an "s". Unless it's the other Steve, in which case he's Olshansky. Simples. ;^)
Very interesting comments from readers, too. I think the core point is this: when you use blockchain/DLT as the basis for a cryptocurrency (as per Charlie Clark's comment), you don't need any external source of trust. It's analogous to paying cash for something: you don't need to trust the person handing you the $5, you just need to trust the $5. The integrity of the DLT and the mining process are all you need.
If you want to use DLT as the basis for trustworthy assertions of something else (such as Identity), then the roots of trust in those assertions have to be something other than the distributed ledger itself. As Brangdon says, "you can tag accounts with certificates to show they have been validated to some level"... but the validation process, and the validating entity, have to be trustworthy on the basis of something that isn't the DLT.
(RIch 27 makes an interesting point, too, about passive linkability vs explicit authentication, but that's the subject of a whole other essay ... ;^) )