You just turn around at the half way point

At which point shielding gets even more complicated.

Less complicated. The rocket exhaust literally forms a plasma shield in front of the decelerating rocket.

Well then, it's easy: Just always face the exhaust in that direction! It's not rocket science!

> always face the exhaust in that direction! It's not rocket science!

You don't see a problem with trying to accelerate with two exhaust streams at 180º to each other?

What if the front one is only a small fraction of the thrust at the rear? Would that be enough to "clear the path". Surely it would cost some efficiency but it may beat he alternative

Whipple shields [1] don't weigh much, and don't involve negative acceleration of any kind.

1: https://en.wikipedia.org/wiki/Whipple_shield

In other words the space-equivalent of a supercavitating torpedo like the Shkval [1], more or less an underwater rocket which creates its own gas bubble around the projectile by deflecting part of the exhaust stream to nozzles on the front of the device.

[1] https://en.wikipedia.org/wiki/VA-111_Shkval

In a universe where every action has an equal and opposite reaction, why not make those reactions argue about it?

what about for the period of time that you're travelling sideways with a largely unshielded broadside facing not exactly empty space?

You wouldn't literally turn around.

You'd probably move your rocket engine over to the other side (or have both a front and back engine in the first place).

Most of the mass of your spaceship will be fuel (like 90%+). You can use that as shielding.

This is already an unrealistic thought experiment, since continuous 1g acceleration for 20 years is so ridiculous that it requires a magic engine (i.e. consuming most of the ship mass in a matter-antimatter reaction) so if we're at it, we can also assume magic shielding which will also shield the crew from all that antimatter and its annihilation.

It's not a coincidence that the original post relativistic spaceship doesn't even bother considering larger accelerations than 0.1g, since achieving even that is a wild assumption.

Still needs to slow down before turning around. ;)