I learned to sew more than forty years ago, and I’m pretty confident about what I’m doing.  Learning about dress history is a different matter, because much of what we want to know simply isn’t there.   Here’s the approach I base my work on, in the hope that you find it useful.

Don’t expect to find definite answers , and don’t trust anyone who gives them.  The further back in time you look the less evidence there is, and the harder it is to interpret.  Conversely, the more tempting it is to jump to simplistic conclusions.

Assess the evidence on offer.   Books of fashion or costume history are a very mixed lot, but the simple rule is to trust only those which tell you how they know: these give the primary source (the contemporary evidence, be it carved, written, painted or dug out of the ground, and where it is to be found) of what they tell you. Then you can go and check their accuracy.  And if they don’t refer to primary sources – put them back on the shelf.

Keep collecting information of every relevant kind.  The more you look at, the more you can see what’s typical and what’s eccentric.  Always record where and when you found something (see above). 

 Try to be aware of what you don’t know.   Has a text been translated or quoted selectively?  Do you recognise ‘antique’ dress in paintings, or formal styles on effigies?  Beware of getting smug about what you do know – the next piece of evidence may force you to revise it all.

Everything is more complicated than you think.  Dress history is often summed up in simplistic generalisations which ignore trade, politics, national variations and even the huge differences between different centuries.  Any basic process such as cutting out a shirt or knitting a sock can be done in more than one way, and each way is ‘right’ to somebody.

Everything is older than you think.   The human race ten thousand years ago was very much like us, only with less stuff.  Their ingenuity engendered what we have today, so don’t ever disdain ‘primitive’ work.

Experiment a lot.   A practical experiment isn’t a scholarly proof, but it can give a working method and it can help us to understand the evidence.  If the garment you’ve made hangs and moves like those in the pictures you’ve based it on you are probably on the right track – and if it doesn’t, you’ve still learned something even if you have nothing to wear. 

Resist the temptation to correct other people in public. It’s bad manners, and you may end up looking foolish.