What is a PhD?

This is the first in a series of articles that I have always wanted to write. I just needed to wait for the right time to write it. I think the time is right – I just completed (defended and deposited) my PhD a few weeks back, so I feel that I am now in a better position to opine about this topic. These are all thoughts that I have always held dear but never wrote down.

What, specifically, is the topic of this post? It is about the things that no one tells you about what a PhD is. This is just a natural side effect from fact that unless you grow up in a campus town, not many people you know will have a PhD.

A PhD is just another degree, much like your Bachelor’s and Master’s degrees. A Bachelor’s degree is the most structured of them, much like a high school diploma. To earn a Bachelor’s you take the required courses and get decent grades in them. A Master’s is less structured. You need to take some courses and perhaps work on a large project. A PhD is the least structured (it is, thankfully, becoming more structured now). You take more classes and then you need to come up with a specific area to work on. Then you need to demonstrate to a group of at least four other professors that your topic has valuable intellectual merit.

Many people have a romanticized view of a PhD. Initially, I shared this view too. We think that PhDs are the world’s smartest people who can solve any problem given to them. This is not really true. Instead of using the term smartest, it is more appropriate and accurate to describe PhDs as reasonably smart people who have persevered to become experts in specialized fields under an academic environment. The keywords are “experts”, “specialized” and “academic”.

In this post, I will clarify why I have chosen those words. And, finally, I will explain why I chose the phrase “reasonably smart”.

Expert. This implies that you are now one of the leading authorities on the topic of your PhD. It means that you have worked long and hard on a subject and understand most of the subtleties behind it. The subtleties are things that they don’t teach you in a textbook because they are too messy to model and explain to a beginner. The subtleties are the things that most people sweep under the rug because it is simply too hard to explain it until you have worked on the topic for a sufficiently long time.

In fact, if you give them enough time, PhDs can talk about all the subtleties of their particular topic – way more than any normal person would care to listen to.

Specialized. This keyword is very important and needs to be combined with the previous word, i.e., expert. For instance, when you complete a PhD in Computer Science, it does not mean that you are an expert in Computer Science. In fact, even if your topic is in, for instance, Parallel Programming, it does not necessarily mean that you are an expert in it. You need to drill down more. For instance, while my topic was in parallel programming, it focused deeply on semi-automated analyses and transformations for a particular style of parallel programming. That is the specialized field that I am an expert on.

Academic. This is another very important keyword. Remember what I wrote about earlier about demonstrating intellectual merit to a group of four professors? Well, turns out that there is a preferred way of demonstrating intellectual merit in an academic setting.

First, let’s see how things work in the real world. If you were in charge of a large project in the real world, you demonstrate competency to your employers by building a complete, reliable and working system that others can use. You are rewarded for making something practical that works.

This is not how it actually works in an academic setting. You demonstrate competency by building a partial and mostly-working prototype that is sufficient for proving what you claim it does. It does not need to be working perfectly. In an academic setting, there is a threshold, above which, doing extra doesn’t get you any other appreciable recognition from your peers. Thus, improving your prototype into a complete working system is usually not seen as a profitable way to spend your time. Your time is better spent working on building the next prototype to demonstrate your future claims.

The unfortunate consequence of this is that sometimes that prototype that you build could be rather detached from the real world. In extreme cases, since you are the only one working on it, it might not even work for other people. Building something that is practical is a good thing to do but it is not pivotal to getting a PhD. I’ve read enough papers where the proposed technique was so complicated that most people would shy away from using it in real life.

So those are the three main keywords. Now let’s talk about the phrase “reasonably smart”. Getting a PhD is more about perseverance than intellect. You do not need to be the smartest person, you need to be the one who is willing to persevere the most. This means investing time on your topic: reading published papers, building prototypes, discussing your ideas, publishing your own papers, etc. Some of these activities can get very tedious and repetitive over time. So the ones who eventually end up completing a PhD are the ones who are steadfast in what they are doing.

Hopefully this post has helped clarified what a PhD actually is. Like I mentioned, unless you frequently interact with PhDs, it is not immediately obvious what it means to have a PhD. Most technical interviewers have the wrong idea of what a PhD is and, thus, tend to have wrong expectations during the interview process.