Unbiased Language

Thinking outside the box is a top skill. It’s the key to problem-solving and enables innovation. But thinking outside the box is also a rare skill because for us to reassemble ideas and find solutions, we first have to disentangle the problem.

And what is the first step to disentangling a problem? It’s unbiased language.

But before taking a look at unbiased language, let me explain how language and bias are intertwined.

Being biased means having a distorted view on a situation. Biases work like filters, and by filtering what we see, they influence how we think and act.

Language, our genuinely human skill, is one of the major filters that shape our thinking. The words we have to describe the world are likely to shape the way we see the world. Edward Sapir claimed about a century ago that language is a perceptive bias. His main argument was that each language has its own set of color terms, which determines the colors we are able to see and name. His alumnus Benjamin Whorf later attenuated the original claim because we actually are able to think beyond our language, or think outside the box, and we’re also able to modify our language if need be.

This means that we aren’t victims of our bias. We can overcome biases with training. We can learn how to see the full picture, or how to see the picture differently. For example, when we learn a second language, we learn that there are different ways of describing the world. Specific words that are present in one language but not in another, like Danish “hygge” (roughly meaning “coziness”), Portuguese “saudade” (roughly meaning “longing”), or German “Schadenfreude” (roughly meaning “mischievousness”), show that a new language can teach us a new way of looking at things.

When we acknowledge that there’s more to a situation than we previously assumed, we are ready to remove our bias. We get the unfiltered gaze. That’s the moment we’re ready to think outside the box.

“Dude! This is unbelievable, the crocodile has thrown in the towel!”

The plain, unbiased message behind Jim’s words is this: Jim’s boss has resigned. Jim’s message, however, simultaneously conveys a few additional messages: 1) Jim and you are on informal terms with each other, 2) you are in the know about the situation, 3) he’s positively excited about the news, and 4) there have been a couple of stressful preliminary events that induced Jim’s boss to resign.

Why do we know this? Because Jim uses an informal word to address you (“dude”), an evaluative adjective (“unbelievable”), an obscure insult that you understand (“crocodile”), an idiomatic expression whose meaning implies there was a prehistory (“throw in the towel”), and exclamation marks. All these elements convey this additional information and add color, or bias, to the original message. And that’s the wonderful feature of language – it allows us to transmit information on content, relationships, and feelings all at once!

The same happens when we describe things as “amazing”, events as “unfortunate”, or behavior as “whining”. Such words are inherently colored. They make language such a creative and fun tool to play and tell engaging stories with. This is how language connects people and thoughts.

Sometimes, however, language should be unbiased. It should come with no filter.

What does this mean?

Unbiased language is an objective kind of language that is free from any color or emotion. It does not attach any emotional value to the content it conveys, be it ideological, political, or any other feeling or attitude.

And given that we’re humans, such attitudes are pervasive. For example, we can present a manager as “smart” or an actor as “glamorous”, or we can do the opposite by saying they’re “oafish” or “hysterical”. This is a way of intentionally and overtly adding our own view to the message we’re conveying. And we do it all the time. We do it when we summarize the news, describe a place or event, or rant about Jim’s boss.

We may also do this covertly. If we use an expression with a positive meaning, we can also subconsciously shape the way it is seen. A “cookie”, for example, is undoubtedly a positive thing. Most of us love them and probably also associate them with family gatherings. If we call tracking data “cookies”, however, users may not fully understand what the implications are and simply assume it’s a good thing. This is how the perception of things can deliberately be shaped by the language we use. This is biased language.

If we do this intentionally, it’s a form of manipulation. Expect this to happen in politics and in marketing – and in fact, politics and marketing are everywhere.

So yes, there’s deliberate manipulation. However, a lot of biased language is not used to deliberately manipulate. Very often, biased language is unintentionally biased. When we express our happiness or sadness about a situation, like Jim’s joy about his boss’ resignation, we automatically use biased language. Otherwise, the message would come across as utterly unemotional and robotic.

So, what is the benefit of unbiased language?

The benefit is the fact that when a situation is presented without bias, the audience can evaluate it independently irrespective of the speaker’s thoughts. Because unbiased language requires the audience to evaluate the message themselves.

This concerns facts and problems as much as it concerns people and ideas. When emotions and opinions are added to a message, they shape the way the message is understood and anticipate the conclusion the audience should come to. Jim’s message is not only a message to inform you about the news, it’s also an expression of him wanting to share his joy with you. If Jim tells you his news in an unbiased version, like “my boss has resigned”, you will have to first think and assess the situation yourself.

In other words, biased language guides the reception of a message while unbiased language favors individual conclusions. And such individual conclusions are the key to new, innovative ideas.

Yet, given that we are humans and love emotional storytelling and bonding, we do not naturally speak in unbiased language. We have to be trained to use unbiased language.

So, when do we actually need unbiased language?

We need unbiased language when factual information is required to support important decisions and develop knowledge. There are professions and sectors in which unbiased language in reporting is a must. They include all the professions in which cases, processes, technologies, or developments are documented. Such reports must contain all the information needed to allow all the involved parties to do their job.

In social institutions, for example, reports must be unbiased because official bodies must take informed decisions based on the information provided. The same applies to medical and legal reports. And, of course, to scientific and technical reports. They must be unbiased because they document the knowledge that was generated and will be used in future decisions, research, and applications.

By using unbiased language, a professional report separates facts from opinions. It may give an opinion or recommendation at the end, but this is unmistakably marked as such. If we don’t remove the bias, the report will produce ensuing biases, like a misleading emotional bias, a methodological bias that overrepresents a specific category in the data, or an algorithmic bias when the machine multiplies imbalanced information because it wasn’t fed properly.

Unbiased reporting means maximal transparency. And transparency enables a fresh, open-minded view on a situation. It is the first step out of the box.

Of course, even if language is entirely unbiased, the content may still be biased. What a report informs about or not makes a difference, and so does the emphasis that is given to specific details. Do newspapers decide to put a law that was passed or a royal family’s feud on their front page? The choice will inevitably influence the discussions people will have on that specific day. The same happens when a scientific paper gives more attention to a small proportion of the big bulk. Such choices add a filter even if the language is unbiased.

But as I said, biased language is not bad per se. We’re humans, so we all appreciate some padding or encouragement when we speak with our beloved ones or when we’re about to hear bad news. Connecting and storytelling are the basis of human collaboration and learning.

The ideal case is the skill of deliberately switching between the two.