But when people talk about whether lockdowns are good or bad they tend to be assuming compliance and considering the effects of that compliance, so I think your definition sows confusion.
Compliance rate is a very commonly considered variable in evidence based medicine - there is nothing confusing about considering it.
Compliance very much varies between intervention types and cultures. Also not remotely controversial.
Assuming compliance is almost always an error in considering the outcomes of medical interventions.
If we are going to reason about what kinds of coronavirus interventions are effective, we must consider compliance.
Whether something is mandated by law or not is definitely a variable in determining compliance.
If it were not, and as you claim, Swedes have been complying this whole time, why would Sweden have started using force of law recently?
Presumably they think it will increase compliance.
Here's a typical narrative that illustrates why we should be more careful about the word "lockdown":
Alex: "Sweden has successfully managed COVID19 without a lockdown."
Bob: "Great, so we should just go and live our normal lives then."
Alex: "Actually no. We need to cut off most social contact, ban visitors to rest homes, and work from home when possible."
Bob: "Er, that sounds like a lockdown to me!"
Alex: "WELL ACTUALLY it's not a lockdown as long as the government doesn't assert the legal authority to enforce these restrictions."
Bob: "... but the effects on health, well-being and the economy will be the same as a lockdown?"
Alex: "If everyone complies, which we expect them to, then yes, but the important thing is that we don't call it a lockdown."
Alex: "Sweden has successfully managed COVID19 without a lockdown."
Bob: "How did they manage that?"
Alex: "They gave their citizens a bunch of guidance on how to reduce the spread of the disease, and asked them to be responsible and most Swedes complied."
Bob: "Why can’t we do that?"
Alex: "Good question - something to do with our government or our culture?"
Bob: "Seems like our government might have burned less goodwill if it had used Sweden’s approach and we’d be in better shape going into this second wave.”
Alex: “Yeah, maybe you are right.”
That means defining lockdown as "government restrictions legally enforced" doesn't make sense. Defining it as "government restrictions widely complied with" would make a lot more sense. And by that definition, Sweden has apparently had a lockdown.
It’s a clear distinction.
That why many people are saying ‘Sweden didn’t have a lockdown’.
What they mean is ‘Sweden didn’t have a legally mandated set of restrictions’.
This is just a true statement.
The fact that the Swedish government made recommendations that Swedes followed without using force of law is interesting, and informative.
True for some people, not for others. I have talked to many people who have repeated "Sweden didn't have a lockdown" and think that means Swedes didn't change their behaviour significantly. They are surprised, even disbelieving, when I point out the quite significant restrictions most Swedes have complied with. These people were misled by news reports using the term "lockdown" the way you do.
Whether or not defining the term "lockdown" differently is the solution, there needs to be some improvement in the way we write about this.
https://news.ycombinator.com/item?id=25119518
Which would imply no lockdown by any definition.