Everybody knows the country is divided. What's not so well appreciated is that there is also an implicit unspoken division about how we conceive the very nature of politics.

The dominant popular conception is that politics is much like ordering something from Amazon. You say what you want, expect it to be delivered, and get the hump when it isn't. That leads to demands that politicians "just get on with it", to talk of a betrayal of democracy, or even to demands for a "strong man" to sort things out.

There is, though, another conception. It's that politics is not just another retail experience. It is about the conflict of interest between people: Robinson Crusoe had no politics until Man Friday pitched up. Such conflicts arise simply because we want different things. The essence of politics is to understand how to cope with this. Sometimes, others' wants are an immutable obstacle to our own desires, so we cannot get what we want even if there is a democratic mandate for it.

At other times, though, their wants might be malleable. This might be done by persuasion, rational or not. But it can be done in other ways. Thatcher, for example, created a constituency of people with an interest in house price inflation and hostility to property taxes by selling off council houses. Blair's expansion of universities created a more cosmopolitan, liberal cohort. And Cameron's great achievement was to take an issue which few (pdf) people cared much about and push it so far up the agenda that it now dominates politics and has changed our political identities.

Radio 4's Today programme gave us (inadvertently) a nice example of this difference on Friday (2'13" in). Martha Kearney claimed there was "no appetite" in Scotland for an immediate new referendum on independence. Nicola Sturgeon replied that "part of the job of leadership is to set out to people why you think change is needed." Kearney was using the politics-as-Amazon conception, in which politicians fulfil orders; Sturgeon was using my second conception, of politics as using persuasion to manage conflict.

Such persuasion, however, cannot always be done by politicians alone. It's a job for all of us. A good example of this is the Northern Ireland peace process. The Good Friday Agreement is perhaps the greatest example in modern times of successful delivery by politicians. And yet the architects of that success have always spoken of the peace process. And for good reason. They know that peace isn't something that can be delivered from on high by politicians even if the vast majority of people want it. It needs constant work on the ground to persuade potential terrorists to abjure violence. The murder of Lyra McKee was a failure of this process. But the failure wasn't one of politicians or police. And it didn't happen when the shot was fired. It happened when someone went looking for a gun and was met with the words "I can sort you out" rather than "don't be a twat sonny."

Now, it's easy for us Remainers to sneer at Brexiters for wanting something that is ruled out by others' preferences. But the silliness is not theirs alone. Many Remainers are demanding a third referendum but giving us little clue that they'll do a less piss-poor job of campaigning than they did in 2016. They care too much for their own desires and pay too little heed to how to influence others.

In fact, this point broadens. There are other policies we might like but which might be prevented by others' behaviour. For example:

- Higher taxes on the rich or on companies. Whether these can actually raise much more revenue or not depends in large part upon whether the rich regard taxes as a price to pay for living in a civilized country or just a burdensome impost which they seek to avoid. This tax morale varies from time to time and place to place.

- Wage-led growth. Whether this succeeds depends upon how companies respond to higher wages. Do they betoken increased aggregate demand and hence a reason to invest? Or are they instead a threat to profit margins and hence a reason to cut investment? Again, the answer will vary from time to time and place to place.

These two policies are much like Brexit, in the sense that whether they can be delivered or not is not merely a matter of a democratic mandate. Their feasibility depends upon others' beliefs and actions. Sensible politics considers what these beliefs might be, and how to shape or change them. Stupid politics, by contrast, acts like a five-year-old solipsistically screaming "I want, I want".

The converse is also true. There's a strong case for some policies even though the voters don't seem to want them. For me, the case for encouraging worker coops falls into this category: they are necessary to increase equality and productivity, whether voters want them or not.

If we lived in an intelligent polity, my point here would be obvious. It's that politics is not and cannot be another Amazon fulfilment centre. It's not just about what you want, but about what others want too. The very essence of politics is about how we manage conflicts between different attitudes, and how we might change those attitudes - the latter being a job not just for politicians but for everybody.