June 2025

Literature Reviews as Discourse

No, this is not a research blog post where I analyze literature reviews and try to reveal their discourses. This blog post is about making sense of difficulties I have had getting published when reviewers are quite critical of your background sections.

Throughout my time as a PhD student, I have been told a great many things about literature reviews, such as:

Literature reviews are summaries.
Literature reviews are about finding gaps in the literature.
Literature reviews are about appealing to certain publication venues.
Literature reviews are about deciding where the field should go.
Literature reviews are about saying which literature we should listen to.

Perhaps the most harrowing, that literature reviews are something you should already know how to do when you arrive for a PhD program.

I’m not saying any of these perspective are wrong. I am sure some people reading can see the mappings to more specific methodological techniques, such as meta analyses or scoping reviews. That’s not the point here, but a useful point nonetheless.

For literature reviews used in the background section of publications, for instance, we don’t really get to understand the thought process or methodology behind them. Our advisors typically don’t get much further than “summarize what’s been said about these constructs or ideas.” Unfortunately, what I have learned through the process of publication and interacting with reviewers and publishers is that to be successful, you can’t just summarize or explain the core concepts used in your article. Saying what has been said before isn’t enough.

Beneath the critique of many reviewers is a disciplinary training that – whether they acknowledge it or not – deeply influences the believability of your literature review. If a reviewer has ever questioned the validity of who or what you are citing, from where it was published or why you aren’t citing from other journals, this may be the reason why. This is especially true when you color outside the bounds of your discipline. It is a matter of epistemology – what counts as knowledge – and methodology – how we prove that it is so. This is perhaps most pronounced if we look to the poles of Western academia, Humanities and Physical Sciences. Where both what counts as knowledge and how we come to “know” seem irreparably distant. Though, these kinds of divides aren’t particularly relevant for most scholars, instead I want to focus on social constructs as a case study into Literature Reviews as Discourse.

What is a “social construct”?

Again, some grace with definitions. In social science, we typically identify (inductive) and test (deductive) theoretical ideas. In this case, I am defining theory as definitions of social/behavioral phenomena that reveal rules about the social world. In simpler terms, theory allows us to know something general about how humans do things, which also lets us *predict* about future human behavior. For example, in social norm theory we are told that humans are highly influenced by their observation of other humans, so the emulation or reproduction of behavior can be expected – especially in new environments or contexts.

This happened on my flight a few days ago. I was in the second group of a Southwest flight and the exit row was completely empty, yet seats before and after were filling up. The first people to board either didn’t want those seats or didn’t know if they were available, so they continued on. This created a bind for the next passengers who had more visual information that they *could not* sit there than that they could, so they passed them up. The normative effect was so intense that the last two people to board sat next to me in the exit row.

Yet, we know based on *my behavior* that theory doesn’t hold for everyone, nor does it hold in every space, time, or place.

This is the balance in social science. We need folks to come up with theory and those to test the boundaries of said theory. So, theoretical or social constructs can be called theory (“social norm theory”) or as a construct (race, gender, sexuality, etc.). Regardless, in Western social science we give them a name and we define them.

What’s the problem with social constructs?

Well, we don’t agree on them. There is a larger argument that is onto-epistemological (what is real *and* what is knowledge, which they are obviously interconnected), which I won’t get into. Suffice it to say that if I coin “polarization” that a number of things will happen with any level of popularity of my term.

Scholars will misinterpret. Scholars will not read the context. Scholars will based their studies on studies that reference your paper, instead of the original. Scholars will theorize a concept and name it the same thing as yours unknowingly. Scholars will disagree with your theorization and say that theirs is the correct version.

And so on. When you dig into the literature review of a concept or construct or theory, you should be getting quite confused. With dissenting opinions, competing definitions, and similar ideas outside of your discipline, what are you to do?

I have heard a couple of answers to this question.

First, latch onto the normative definition. It will make everything easier and there is a more

Second,

Third,

On With the Discourse

This is what I mean by “discourse.” The validity and utility of knowledge, especially theory, is subject to political contestation. If you reflect on your training as an academic this should make sense pretty quickly. If you are in statistics, you are trained rigorously to understand uncertainty and you employ or create models that assess certainty. It makes it easy to read claims in Literatures in English and have a hard time feeling their knowledge is valid. Likewise, if you are a science and technology scholar, it is hard to validate machine learning research in our current era as valid.

It’s not to say there is one discipline that is “correct”, instead what I am trying to reveal is that there are a few crucial questions you should ask yourself when conducting a literature review:

1. Am I trying to find *my* truth in the literature, or am I reproducing the norm?
3. How much effort and time am I willing to invest in the publication timeline?
2. Who am I trying to appeal to in order to get published?

I am not judging those who want to comply with a journal’s norms, or truly believe that the most well-cited papers are the best papers. What is important is that when reviewers are reading your work there is a good chance they are familiar with the normative, popular lines of logic, and not with critical or dissenting ones.

If your background section is non-normative, then it can cast your entire study under doubt.

Instead of just defending your academic contribution, now your priors are under scrutiny as well. If we don’t agree or understand how you are building the foundation for your study, naturally the validity of the method and findings are more difficult to approve of. Whereas, if you have a background section that looks like everyone else’s, reviewers can glaze over and begin to assess your contribution. There is much less friction and fewer moving pieces this way.

Publication is an inevitable compromise. It will knock you down, drag out the process for months on end, be unfair, be helpful, be unhelpful, and by the end you just want it to be over. This is why sections of publications seems odd or out-of-place. An extra study or defensive tone. Unfortunately, research articles often have very little to do with the people that read them and instead prioritize the team that wrote the study and the editors/reviewers that gate kept it reaching the academic community.

This isn’t necessarily a bad thing. We can’t, or shouldn’t, publish in a vacuum. Our peers are overworked and underpaid, so having a Reviewer 2 every once in awhile is going to happen. However, it is naive to enter this process with the ideal of rigorous academic work as the only bar you must overcome to get published. Generally, you will either have to spend extra time and resources making your way valid to the review committee (which you could not reach that opportunity with a desk reject) or you comply with journal norms and ease the review process. Hopefully something in between leaning towards the former.

Hope this helps!