How to find research topics and publish more?

You're thinking about this the wrong way; if do not have a topic you want to write about - then don't write.

I'm not being flippant: You should write and publish something if you feel it's important to express in writing and to disseminate. Don't try to produce publications for publication's sake.

But maybe it's the other way around: You're frustrated about being so passive - like you said, just "reading, reading, reading" - rather than producing something written. Well, in that case, instead of thinking about what to write, think about what you want to know or figure out. What questions about human society (be it history, sociology, psychology or whatever you're focused on) intrigue you?

When you have your questions, you look for what's already been written or said. If it's sufficient - then you had an interesting read and can perhaps put the knowledge to some use; and if there isn't a definite answer to your question - that's what you can contemplate, explore, study, and eventually publish about yourself.

Edit: I gave a general answer, but your tags suggest you are in grad school. If that's the case, then also consider:

• You have an advisor. Talk to him/her about this - that's what s/he's there for.
• If you've submitted a research proposal, then you've asked yourself some of these research questions already. Are you actually looking into them? Do you have satisfying answers to them?

I am in a different field and I am almost in a similar situation where I am interested to work on something original. However, getting started is bit difficult at this point.

As pointed out by @einpoklum in their answer, 'You should write and publish something if you feel it's important to express and disseminate. Don't publish for publication's sake.' This seems to be a great piece of advice.

In my field (engineering), we do read a lot of papers; but, I read it differently.

Case A:

• Read a paper that reports a new method M to solve a problem P.
• Find out the limitations of the method when P changes a bit.
• Can method M be improved to fit the new problem P'.

Case B:

• Read a experimental paper.
• Find out the outcomes of the experiments and all its parameters.
• What happens the method in experiment if the problem changes? (a vice-versa to Case A)

Case C:

• Start exploring a problem which has never been solved (an open problem).
• Do there exist some approximation methods for this problem? What are their results?
• Can it be solved in an alternative way?

I am really not sure whether it would fit to humanity field.