⚠️ Warning ⚠️

This article contains strong language that some readers may consider offensive.

Whether you admit it or not, taboo words are an integral part of our lives. We say them, whisper them, and shout them. As Mark Twain once said:

Under certain circumstances, profanity provides a relief denied even to prayer.

Scientists discovered that swearing like a sailor even helps tolerate physical pain, relax and even bond. Yet, despite the positive effects swearing does bring to the table, naughty words are still naughty. And in business communication, taboo language is an absolute no-no-no.

Emails sent with cussy words don’t only demonstrate your professional level, but actually can be considered as bullying by opponents.

And meanwhile f*ck, $#@*% and b*tch are well-known, there are too many not so obvious representatives of offensive words, which may spoil the communication picture. For instance, seemingly harmless, amateur, chin, girl or pot may also sound profane. 

As you have already guessed, today we’ll be talking about how text checkers can help you deal with colorful language.

The variety of swear words: Schimpfworte, gros mots, palabras rudas

Every language (maybe except for Esperanto) has a dark side with rude swear words, insults, hate speech, profanity, vulgarisms, etc. Oh, Esperanto isn’t an exception. Sorry, but Wiki says it does have vulgar words too!  Believe it or not, swear words have a long history; and interestingly they weren’t always considered as taboos. Just dig more about sh*t or damn.

But that’s not the point.

Swear words, profanity, and insults can be hidden behind ordinary words, and if so, we need to filter them in communication. These words can be grouped in different categories, e.g. general swear words, sexual references, religious insults, race and ethnicity insults, etc.

Explore more

WProofreader browser extension

Free for individual users and 14 days-trial for business users.

Explore more

Profanity has historically been considered punishable. 

In the US, profanity can be regulated. However, under certain circumstances, it is consistent with the First Amendment.  For example, federal law prohibits obscene, indecent and profane content from being broadcast on the radio or TV. If not legally banned, bad speech is absolutely inappropriate in business communication. That’s why the demand for proofreading solutions with a profanity filter is increasing.

In different countries, practices vary. In France, e.g. there’s a law against hateful content with gros mots on the internet, in Germany — Netzwerkdurchsetzungsgesetz which responds to the increasing spread of hate crime and other criminal content incl. Schimpfworte. On the contrary, the Spanish Parliament removed the blasphemy law from their legal system and moved closer to bolstering freedom of expression and palabras rudas.

By and large, for many niches and companies, profanity in business communication isn’t acceptable and may impact brand reputation, image, trust and credibility, even lead to lawsuits. Sentence checkers may be helpful to an end.

But, how effective are they in combating profanity?

Comparing profanity checkers on bad speech examples

☝️An important update.

In the latest product release, we introduced custom rules for non-inclusive and profanity writing in English. When you write a non-gender inclusive term, e.g. businessman or businesswomanWProofreader marks it and gently suggests unbiased alternatives like business person.

Another case is swear words. Profanity is an absolute no-no-no (ok, sometimes there may be solid reasons). So, we’ve also created custom rules to detect profanity in writing.

Try Style guide now

For the purpose of this article, we’ve gone deep and analyzed the performance of popular sentence checkers on a variety of offensive words in English, German, French and Spanish. Frankly, the task wasn’t an easy one, yet we learned many interesting things and new words 😉

Here are insights we want to share with you.

English

1,955 offensive cases were derived from the free sources like Wikipedia, GitHub repositories, thematic blog posts and articles.

We’ve found out that there are many cases when words or phrases with a general meaning can be considered as offensive words (also insults or profanity), e.g. explosion, bomb, slave, etc. Besides, English has many offensive words and phrases derived from foreign languages.

Grammarly
Grammarly

Grammarly and ProWritingAid (free versions) detect some cases as misspellings. However, they don’t give any alerts or suggestions. Sapling AI goes further, and warmly invites you to replace misspelled profanity words with those written correctly, e.g. fcukfuck.

Sapling AI

German

German curse words can be split into Fluchwörter, Schimpfwörter, Vulgarismen, Blasphemie, and Diskriminierung.

Fluchwörter are rude expressions used to show vivid emotions, for example, anger, disappointment or a wish for something bad to happen. For instance, Fahr zur Hölle!.

Schimpfwörter endow any object (person or object) with negative properties, thereby offending or humiliating him. They are often formed from words such as “shit” (and everything associated with it) and all kinds of names for body parts, as well as animals. Germans say Ah du Schwein! when they’re angry or irritated. 

Vulgarismen are a combo of two previous groups; words used in everyday speech to offend someone. For example, Scheiß drauf! — Screw it! Or Verdammte Scheiße!

Many curse words are blasphemies at the same time. 

Diskriminierung or discrimination refers to disadvantage or degradation of groups or individuals based on certain values ​​or due to unreflected, e.g. T. also unconscious attitudes, prejudices or emotional associations.

We’ve checked the German dataset of 2638 cases with Duden Mentor, probably the most famous German text checker.

Duden Mentor
Duden Mentor

Some cases are underlined with suggestions provided, some come with no suggestions. The top offensive words such as Hure, ficker, Scheiße are not detected as misspellings. Duden Mentor doesn’t recognize the majority of these profanity words and consider them as misspellings.

French

French is one of the most expletive languages. There are about seven categories of jargon incl. argot de la polic and argot de la drogue. Yes, Frenchmen do like swearing and the rude words are even allowed in the official media.

For the research purpose, we’ve collected 4,282 cases in the dataset and checked them with LanguageTool French, and Outwrite French. Both checkers underline about 20% of cases as misspellings. Importantly, such key swear words as merde, connard, putain, and zut aren’t underlined, which means they’re in the checkers’ dictionaries.

LanguageTool
LanguageTool

Spanish

It’s probably the strongest language of all! Spanish people do fancy seasoning their speech with sharp words and we can’t really fault them for that. So, our huge dataset includes 4117 cases, in particular, different variations of mierda, joder, and puta. Correcto., a widely-used Spanish text checker, doesn’t detect top offensive words as misspellings, which means they’re present in its dictionaries or linguistic models. Yet, some profanity cases are marked as misspellings.

Correcto.
Correcto.

Summary

Businesses that use strong words in their communication are needlessly damaging their brands. Profanity isn’t always easily recognizable, especially for non-native speakers. So, you may never know whether you sound respectful or abusive when talking to a teammate, client or business partner from another country. 

To clean up your writing, you will probably need a writing assistant. But can sentence checkers cope with profanity? Well, wrapping up the above — as for now, NO.

To clean up your writing, you will probably need a writing assistant. But can profanity checkers online cope with this task? 

Well, wrapping up the above, the majority of bad words filters don’t have a dedicated feature “profanity check” except for WProofreader Style guide and the Sapling AI profanity filter. But the latter is offered as a public demo, not integrated functionality. At least, for now.

Grammarly and ProWritingAid probably have exclude-lists with well-known offensive words as they do not give suggestions for English cases like fck, beatch, etc. However, this is not a separate feature, and text checkers don’t mark these cases as profanity. In the majority of cases — it’s an unknown word or a misspelling. 

As for other languages, writing assistants either ignore profanity or detect such words as errors.

“Profanity checkers or filters”, you name them, are likely to become a highly demanded feature in the nearest future. There are several opportunities for their tech implementation: exclude-lists, profanity alerts, style improvement or a more advanced one, the rewriting feature. WebSpellChecker keeps working in this direction now; we’re planning to extend this feature to other languages soon. So, stay tuned.