Beware noun phrases


Co-written with Matthew Jordan; cross-posted from Curing the Curse. Published 2020-08-27.

Hello and welcome to lesson four of Curing the Curse of Knowledge! In the previous lesson, we said that the Curse of Knowledge causes writers to think in abstractions, which leads them to write top-down explanations when bottom-up explanations would be more clear. In this lesson, we want to show you that abstractions don’t just impair writing on the level of documents and paragraphs. They also wreak havoc on sentences.


Let’s begin with an example. Here is the beginning of a blurb for a talk at a mathematical biology conference. Fair warning—it’s a doozy.

Biological populations facing severe environmental change must adapt in order to avoid extinction. This so-called “evolutionary rescue” scenario is relevant to many applied problems, including pathogen evolution of drug resistance during the treatment of infectious diseases. Understanding what drives the rescue process gives rise to interesting mathematical modelling challenges arising from two key features: demographic and evolutionary processes occur on the same timescale, and stochasticity is inherent in the emergence of rare well-adapted mutants.

If you’re like us, reading those three sentences was like wading through molasses. Why are they so hard to read? In short, it’s because the writer went way overboard in their use of this lesson’s central idea: noun phrases.


As you know, a noun describes a person, place, or thing: dentist, the Galapagos, blowtorch. Noun phrases are collections of words that serve as nouns: my dentist’s addiction to laughing gas; my infamous 2003 Galapagos ferry ride; my look of horror upon realizing I couldn’t find the blowtorch’s off switch. There is nothing inherently bad about noun phrases, but when we overuse them, we rob sentences of action. To illustrate, imagine telling a story that goes like this:

My sleazeball uncle bought me a new car. I was initially grateful, but the windows were stained and the colour was off-putting.

This passage is pretty crisp. The reason is that the subjects and objects in each sentence are all either nouns or short noun phrases: “my sleazeball uncle,” “a new car,” “I,” “the windows,” “the colour”.

Enter the Curse of Knowledge. When we become over-familiar with stories—that is, gain expertise about events and concepts—we tend to refer to them instead of recounting them in full. Instead of recounting, “my uncle bought me a car”, you might refer to “my uncle’s gift”. Instead of recounting, “I was grateful”, you might refer to “my gratitude”. But if you overdo it, you might end up with a behemoth like this:

My initial gratitude for my sleazeball uncle’s gift of a used car in an off-putting colour with stains on the windows quickly subsided.

In this sentence, the writer has taken multiple events—“my uncle bought me a car,” “I was grateful,” “the windows were stained,” and “the colour was off-putting”—and described them as a single, abstract entity: “My initial gratitude for my sleazeball uncle’s gift of a used car in an off-putting colour with stains on the windows”. It’s like a deadlift for your brain. By lumping all these events together, the writer forces you to trudge through 21 words before you finally reach the verb, “subsided”, where the action is.


Returning to the mathematical biology blurb, we can now see that it has the same problem: it’s full of long noun phrases that stall the action and impose a heavy cognitive burden. Here is the blurb again, with all the nouns and noun phrases highlighted:

Biological populations facing severe environmental change must adapt in order to avoid extinction. This so-called “evolutionary rescue” scenario is relevant to many applied problems, including pathogen evolution of drug resistance during the treatment of infectious diseases. Understanding what drives the rescue process gives rise to interesting mathematical modelling challenges arising from two key features: demographic and evolutionary processes occur on the same timescale, and stochasticity is inherent in the emergence of rare well-adapted mutants.

75% of the words are highlighted! This writer’s expertise in mathematical biology has caused them to condense processes into noun phrases. The Curse of Knowledge has left its mark; no wonder it’s hard to read. To make it clearer, we should unpack the processes contained in the noun phrases. Here is a revised version:

When biological populations face severe environmental change, they must adapt in order to avoid extinction. This phenomenon is called “evolutionary rescue”, and it happens to viruses when we attack them with antibiotics. Evolutionary rescue is hard to model mathematically for two reasons. The first is that evolutionary rescue happens so fast that we struggle to distinguish between evolutionary changes and demographic ones. The second is that evolutionary rescue involves mutation, which is inherently random.

Now only 40% of the words are highlighted. Not only that, but the nouns and noun phrases are spread apart, allowing the reader to take in the information one bite at a time. Notice how we took the first big noun phrase, “biological populations facing severe environmental change”, and turned it into a clause of its own: “when biological populations face severe environmental change.” To the writer, it’s a difference of just a few keystrokes. But to the reader, it could be the difference between reading the passage and giving up.


For this lesson’s activity, your goal is to highlight the noun phrases in the next part of the biology blurb. If you’re feeling bold, try to unpack them and rewrite the passage. Here it is:

In this talk, I will present recent work on population dynamics in changing environments, merging biological realism with tractable stochastic models. Firstly, I will describe a model of drug resistance evolution in chronic viral infections, which serves as a case study for a novel mathematical approach yielding analytical approximations for the probability of rescue. Secondly, I will present a combined theoretical and experimental investigation of the classical problem of establishment (non-extinction) of new lineages, using antibiotic-resistant bacteria as a model system. Finally, I will discuss some future directions in modelling antibiotic treatment to predict optimal dosing strategies, and in developing a general theoretical framework for evolutionary rescue that unites approaches to distinct applied problems.

Send us your highlighted or rewritten versions, and we’ll send you ours!

The final lesson in Curing the Curse of Knowledge will focus on the single most valuable resource you have as a writer: a trusted editor.


Pages that link to this page