Twitter Linguistics: A Study

Serena [redacted]

Mr. [redacted]

Linguistics A

February 27, 2018

Twitter Linguistics: A Study


        Within the last twenty years, technology has taken the world storm, and as a kid born in the year 2000, I grew up in the midst of this technological boom. I remember spending afternoons playing computer games with my older sister in pre-school, being glued to my Nintendo DS portable gaming system in elementary and middle school, and when I got my first smartphone in high school, spending more and more time online.

        As technology has advanced and become more accessible, more young people are finding themselves in the same situation that I am finding myself in right now--I spend a significantly large portion of my time online. This means that the user-bases for many social media platforms have skyrocketed, therefore creating sizable online communities and making interaction between people from anywhere and everywhere easier than ever.

        Online communication did not come without its hardships, however. Some of the most important traits of clear communication include facial expressions and tone of voice, and both of these traits are removed when communicating through text, whether it be through an e-mail, text message, or a tweet. When one reads the thoughts and feelings of another, he or she interprets it in his own and has to make an assumption about the tone of the speaker and how the speaker looks while communicating.

        To compensate for this loss, internet users have adopted a system of universal internet slang. It began with basic symbols, such as emoticons representing faces like :-) and :-(, and soon grew to include acronyms (LOL for laugh out loud, JK for just kidding, et cetera). Given the sheer magnitude of the online population, this agreed upon internet lingo branched off into different subsections, with each evolving further and further, thus forming linguistic layers, with the most niche internet lingo being found at the deepest layer.

        I personally am an avid user of Twitter, so for my linguistics project, I wanted to analyze some of the internet convention linguistic phenomena I have commonly observed from both friends and strangers on Twitter, specifically.  

Experimental Design:

        Although my linguistic research was not based on testing a hypothesis of any sort, I designed a survey which I tweeted out to gather sentiment on some of the online linguistic commonalities that I was considering studying and analyzing.

After my survey reached 350 responses, I closed it, as a result of the percentage data getting muddled with user-submitted responses as well as having a lack of time to analyze and consider every single answered free-response question.

My survey consisted of eight sections total (seven different linguistic phenomena/groups of linguistic phenomena one feedback section). The seven linguistic sections I used were:

  1. Keyboard Smashes
  2. Comma Ellipses
  3. Excessively Long Ellipses
  4. Phrase Capitalization and Period Insertion
  5. Non-Interrogative Question Marks
  6. Midway Capitalization
  7. Space insertion

For all of the linguistic phenomena, I included two main types of questions (along with space for elaboration):

  1. Comparison with a formal English alternative.
  2. Perceived frequency of usage.

        For linguistic phenomena that I grouped together, I also included the following types of questions:

  1. Comparison with the other alternative choices of online lingo in the group.
  2. Comparative frequency of usage with online lingo alternative.

        For certain linguistic phenomena that seemed to be based on an arbitrary placement and ordering of certain characters (namely, keyboard smashes and space insertion), I also included the following types of questions:

  1. Whether there were such things as a more appealing version and a less appealing version of the linguistic phenomenon.
  2. Criteria for the formation of a more appealing version of the linguistic phenomenon.

        In my analysis, due to a lack of time, I did not end up using all of the data I collected, and only analyzed keyboard smashes, phrase capitalization and period insertion, excessively long ellipses, as well as comma ellipses. However, to reflect the fact that my analysis is far from a comprehensive analysis of the many linguistic phenomena that have arisen from niche communities on social media, I have still included all sections of my survey in the attached sample responses.


  1. Keyboard Smashes

A keyboard smash, simply put, refers to the random combination of letters that is typed out when a person smashes their fingers on a computer keyboard without thought. The phenomenon has grown especially popular Twitter, a platform where many internet jokes, also known as memes, are birthed. Of course, the natural human instinct when met with something funny is to laugh; thus, keyboard smashing is mainly used as a linguistic tool to emulate laughter, and, in many online communities, has begun to replace other signifiers of laughter, such as acronyms like “LOL,” “LMAO,” as well as the classic “hahaha.”

As keyboard smashes have begun to phase out these signifiers, they have begun to pick up different nuances in meaning as well. In a survey of 350 Twitter users, over 92% of users felt that a keyboard smash in all capital letters represented a larger laugh than a “LOL,” whereas a mere 0.9% felt that “LOL” represented a larger laugh and .9% felt that they did not differ (the remaining percentage of responses consists of custom user-inputted responses submitted by the surveyed user). Users commented that compared to the acronym “LOL” (which feels and looks much more structured and controlled), when someone keyboard smashes, it is “like [he or she has] no words” and “can't contain whatever emotion [he or she is] feel[ing]”—“whatever [he or she] saw was so funny that [they’re] laughing too hard to properly type.” Furthermore, 95.4% of surveyed users felt that a keyboard smash in all capital letters represented a larger laugh than a “haha,” with many users commenting on the negative connotation that “haha” has picked up. Many users used adjectives like fake, weak, emotionless, sarcastic, forced, unamused, bitter, bland, and old-fashioned when describing what kind of laugh “haha” was, with one user making the statement that “if [a person] use[s] “haha” in 2018 [they] sound passive aggressive.” The reasoning behind why the keyboard smash seems to have better, more genuine connotations than “haha” seems to be similar to the reasoning behind why a keyboard smash seems to be a bigger laugh than using an acronym like “LOL” is—“haha” is a controlled, rigid statement, that takes little effort to type but still requires thought. A keyboard smash, on the other hand, is natural, instinctive, and takes little effort to type because it does not require thought to type out.

To delve even deeper, different forms of keyboard smashes have different connotations and usages as well. As mentioned before, keyboard smashes in all uppercase letters are generally accepted as representative of a loud, uncontrollable laugh. This definition of the uppercase smash holds up when compared to other keyboard smashes. 64.9% of users agreed that a keyboard smash in all capital letters represents more of a loud, uncontrollable type of laughter, whereas a keyboard smash in all lowercase letters is used for things that are funny, but not insanely “I-couldn’t-hold-in-my-laughter” funny; essentially, a lowercase key smash is the equivalent of a snort or a chuckle in real life. Logically, this makes sense—capital letters in writing typically indicate an increase in volume or intensity, and loud volume is one key characteristic of a booming, uncontrollable laugh. Since lowercase key smashes are smaller, they are quieter in comparison, and can thus be used to represent sheepish or awkward laughter. For example, if someone was sent a video of a person slipping on ice and thought it was funny despite the fact that the person got hurt, a fitting reply could be something along the lines of “ksdfhsjdfhskdjfh I feel bad for laughing.” The use of lowercase letters is fitting here compared to uppercase letters because the speaker does not want to bring too much attention to the laugh itself; the speaker is restraining himself slightly.

A sizable percentage of users (46.3%) agreed that compared to uppercase key smashing, lowercase key smashing is more fitting to be used as a filler or in place of punctuation regardless of whether the situation is funny or not. Surveyed users commented that “lowercase key smashes tend to substitute punctuation especially when a period feels too formal and other punctuation doesn’t feel right,” and many also commented that lowercase key smashes typically come off as gentler and softer. In the same way that using uppercase key smashes to represent uncontrollable laughter makes sense, the soft connotation of lowercase key smashes is logical as well—lowercase letters represent normal to low volume voices, and physically are smaller and generally rounder than their uppercase counterparts (eg. a is rounder than A), creating both a softer sound-image as well as a softer visual appearance. One user in particular commented that they use lowercase keyboard smashes to soften a sentence because “it makes things seem more friendly, casual and light. Because tone and body language aren't present in online communication, a lot of people are concerned about coming across as too blunt or rude without including small fillers and colloquialisms like these.”

As just seen, visual appearance has grown to become one of the most important aspects of a keyboard smash. In fact, it has become so important, that over 84.6% of surveyed users agreed that there are such things as a "good" keyboard smash and a "bad" keyboard smash. Thus, the question then became, what exactly is the difference between a “good” and a “bad” keyboard smash anyways?

Out of the 84.6% of users previously referred to, 69.1% agreed that the keyboard smash should contain little to no vowels (eg. JGHSDSDHJFS is a better keyboard smash than OUIUSIUSFI is). This “rule” of keyboard smashing most likely exists for the same reasons why lowercase keyboard smashing comes off as less strong than uppercase keyboard smashing. Vowels make softer sounds, and consonants make harder sounds, and so, for the example of an uppercase keysmash used to convey convulsive laughter, the goal of communicating an uncontrollable, loud, disruptive laugh seems to fit more with consonants, that make harsher and harder sounds. Visually, vowels generally are rounder and less complex shapes than many consonants are, so consonants are more fitting to use for uncontrollable laughter as they tend to add to the visually chaotic and intense nature of the keyboard smash.

64.6% of the 84.6% previously referred to agreed that a “good” keyboard smash should not contain too many spaces. In the example provided in the survey, these users felt that SHFKJSDHFKUJD was better than DFD FS O  DBF D. As spaces tend to make a person mentally pause when reading, this makes sense as breaking the flow of the keyboard smash is less chaotic than a keyboard smash with no spaces. This also reflects the same principles as real life laughter—if someone was laughing out loud but pausing every few seconds, it would be pretty awkward, and not seen as an honest, uncontrollable laugh that the person instinctively let out.

51.4% of the 84.6% previously referred to agreed that a “good” keyboard smash should not contain letters in an obvious order. In the example given in the survey, these users felt that HFISUHDFKJSDF, a keyboard smash I randomly banged on the spot, is a better keyboard smash than ASDFGHJKL, all of the letters in the middle row of a QWERTY keyboard in order, is. This “rule” is most likely attributed to the necessity for a keyboard smash representing loud laughter to be chaoticif the laugh represented by the smash is supposed to be an uncontrollable one, the keyboard smash should be uncontrolled as well.

  1. Tools for Emphasis: Phrase Capitalization, Period Insertion and Excessively Long Ellipses

As mentioned in the introduction, one of the biggest losses in online, non-verbal communication is the loss of tone of voice and intonation, which is used for a variety of things in real life, with one of the biggest uses being for emphasis. A way around this when typing text, in some situations, has been to bold, underline, or italicize the phrase, but this is only possible in rich text editors and is not available on a platform like Twitter. Thus, in attempts to compensate for this loss, three main tools for emphasis, with similar (but not identical) connotations in usage, emerged in online communities: phrase capitalization, period insertion, and excessively long ellipses.

The first, more commonly used tactic for emphasis, phrase capitalization, involves capitalizing the first letters of key words or a key phrase in a sentence. Note the difference in the following two sentences:

I’ve been eating snacks for two hours.

I’ve been eating snacks for Two Hours.

As can be seen, in the second sentence, the first letters of the words in the phrase “two hours” have been capitalized. 76.9% of surveyed users felt that the first, unaltered sentence flowed more smoothly than the second one when read over, with most indicating the effect of the disruption while reading as being the emphasis of the length of time the subject had been snacking for (thus insinuating that the length of time had been an extremely long one). This emphasis most likely occurs because capitalization of the first letters of words in the English language typically indicates some sort of importance—proper nouns and titles, for example, follow this same pattern of having the first letters in each word capitalized, whereas common nouns stay completely lowercase. So, in the example sentence above, the phrase “two hours” is given importance over the rest of the sentence as a result of formalities in standard English writing. Moreover, I personally have found that I read capitalized phrases differently out loud, slightly announcing (but again, extremely slightly) a phrase when it is capitalized than compared to when it is lowercase, so, it seems that capitalization tends to mimic that inflection in tone as well.

Broken down further, surveyed users commented that the act of capitalizing a non-proper noun is generally not grammatically correct, and so, in the context of an otherwise correctly capitalized sentence, the incorrect capitalization breaks the reader’s flow of reading and singles out the improper phrase. Having the first letters in “two hours” capitalized also brings visual attention to the phrase in the context of the lowercase rest of the sentence because the capitalized letters are physically bigger and sharper in appearance (eg. the uppercase H in “Hours” is more jagged of a letter than the lowercase h in “hours” is). Thus, visually, the phrase is also emphasized.

The second tactic for emphasis, period insertion, is, although not as commonly used as phrase capitalization, still another well known way to single out certain words or phrases in a sentence. Take the following two sentences:

I’ve been eating snacks for two hours.

I’ve been eating snacks for. two hours.

As can be seen, in the second sentence, a premature period has been inserted before the phrase “two hours.” 89.7% of surveyed users felt that the first, unaltered sentence flowed more smoothly than the second sentence, which is logical, as, when people see a period in a sentence, they instinctively stop or pause before continuing on. Many people felt that the insertion of the period still served as some type of emphasis in the sentence, but 67.4% of users felt that the produced emphasis was different than that produced by phrase capitalization. Most commented that since the period forces the reader to stop unnaturally in the middle of the sentence, a somewhat awkward, uncomfortable environment is created, meaning that the phrase “two hours” more time and space in the environment, causing an awkwardly toned emphasis, as opposed to the more natural, inflective, direct emphasis of the phrase itself produced by phrase capitalization.

Given the difference in perceived tone, many surveyed users felt that the two different types of emphasis projected different connotations on the sentence. Many who indicated this felt that adding a period simply made the sentence more uncomfortable to read (which can also be attributed to the fact that 57.1% of surveyed users felt that period insertion was less commonly used than phrase capitalization and that 8.3% of users had never seen period insertion used compared to the 2.3% that had never seen phrase capitalization used). One user in particular commented on the differences in effect between the two tactics, writing that “capitalization brings more dramatic emphasis to the phrase, but the period denotes the speaker actually being taken aback at just how long they’ve been eating snacks for, as if they themselves had not realised until that very moment.”  Another user wrote, “capitalization is more of a grand (and probably sarcastic) “LOOK AT THIS. The Importance of the sentence!” while the period insertion gives more of a "I'm tired" and a "can you believe this shit?" kind of feeling, like a pause for resigned acceptance or disbelief.” Evidently, the hesitation produced by the premature period being inserted causes the writer to come off as slightly ashamed of (in this example), how long they have been eating snacks for, thus giving the sentence an overall more negative tone—and all of this is achieved through the insertion of a single character.


A third tool for emphasis, which, in my survey, I did not include in the same category as phrase capitalization and period insertion but, in hindsight, I wish I did, is the use of excessively long ellipses. Compare the following sentences:

He left…

He left…………………………………………………………

As expected, the large majority (86%) of surveyed users felt that the second sentence read longer than the first one, which makes sense as not only does the excessively long ellipses appear visually longer, but also, normal ellipses are typically used to trail off, so, when this fact is projected onto an extended ellipses, the trailing off of the sentence when read seems to extend in the same manner. In this sense, the long ellipses represents a longer space of silence after the words in the sentence are read that exists both when there is content following the ellipses and when there is not, thus creating a prolonged, incredulous silence that exists to give time for the content of the sentence to sink in. The thought preceding the ellipses occupies a larger space and a longer period of time, emphasizing it and making it more dramatic—the period of time where nothing is happening is longer, which, through speech and text, communicates the idea that the speaker does not know what to say or how to react. This matches with what many people described the effect of the excessively long ellipses as—a tool that, aside from just representing a pointed silence (eg. if I asked a friend what a book they’ve been wanting to read was to buy them a birthday present and replied to their response with “interesting……………”) can, more interestingly, communicate both the feeling of awkwardness, and more commonly, the feeling of a negative kind of shock because in both situations, the writer does not know what to fill the silence with.

In terms of examples of awkwardness, one surveyed Twitter user wrote that they, in fact, use long ellipses when they “don’t know how to end a tweet.” Another wrote that they commonly used excessively long ellipses with phrases and fillers like “um, okay……..…” and “I mean…………… I guess” to highlight an awkward response to a situation. Essentially, the usage of excessively long ellipses for awkward situations is based on the fact that long, unnecessary silences are, many times, awkward silences, because no one knows how to continue the conversation or thought.

In terms of examples of shock, one surveyed Twitter used wrote that they associate long ellipses with the sentence “I can’t believe this” while another associated it with the sentences “Wow……... you really did that? Yikes.”  One user elaborated on the shock effect, writing that if they typed the phrase “omg…”, it could be in response to something that simply “made [them] gasp,” but if they typed the phrase “omg…………”, then their “physical reaction was probably more drawn out and emphatic than gasping, such as [their] jaw literally dropping and staying open.” When using extended ellipses after a noun (especially in comments on photos), one user wrote that, for example, on a picture of a guy, there is a difference between the comment “his shirt…” and “his shirt……………..”, as the second seems to add emphasis that there is something about the person’s shirt that the writer does not like or finds shocking in a negative way. In shocking situations, the excessively long ellipses evidently represents a dramatic pause of disbelief, to the point where the writer has no words.

It is worth noting that many people wrote in the survey that the use of long ellipses does not fit in serious situations, as it carries a tone of a funny type of disbelief. For example, in the example above where the user wrote that they associated the extended ellipses with the phrase “Wow……... you really did that? Yikes”, the said phrase emphasizes the fact that the person did, indeed, do that, rather than emphasizing the fact that the speaker cannot believe that they did that. The long ellipses thus seems to amplify the stupidity of whatever action the person took, therefore giving the shock of the speaker a funny, unserious tone. For truly serious matters, someone confronting the person that took a problematic action would call for a completely serious tone—the confronter would probably say something much more formal, maybe along the lines of “I am extremely disappointed in you for doing this.” Ellipses would probably not be used at all, as the opinion of the confronter that the perpetrator did something very wrong is a strong one. Likewise, a second example can be seen in the example above that compared the phrases “his shirt…” and “his shirt……………..”. The second phrase with the long ellipses would point to something that made the wearer of the shirt look silly or stupid to the writer—perhaps, if there were rainbow feathers spouting out of the sleeves. Through that phrase, the writer does not emphasize the fact that there is something wrong with the person’s shirt, but rather, that there is something notable on the shirt that the user finds so repulsive that is funny. However, if something extremely offensive was on the persons shirt (for example, a racial slur), the writer would not perceive the shirt as making the wearer look silly—the writer would perceive the wearer as hateful and offensive and confront them on much more serious grounds. Once again, the use of ellipses, excessively long or not, would not match the situation as the beliefs of the writer are strong.

  1. Comma Ellipses

        Another, slightly different, use of altered ellipses is the growing use of comma ellipses. Essentially, comma ellipses are exactly what they sound like—ellipses, but created using repeated commas instead of periods. They are largely used to communicate the writer feeling uneasy or unsure about what they are saying, or to communicate the fact that whatever the writer is saying should not be taken too seriously. Take the two sentences below:

        I don’t know…

        I don’t know,,,,

        As can be seen, the ellipses in the first sentence has been replaced with repeated commas in the second. The majority of surveyed users (over 64%) felt that the second sentence seems more unsure and awkward than the first sentence, which seems to make sense given the differences in traditional uses between the comma and the period. Commas are short pauses compared to the full stops that periods represent—one is a rest in the middle of a thought, and the other ends a thought. Thus, in the concept of ellipses, although both “...” and “,,,” trail off, the period ellipses seems much more certain compared to the comma ellipses—the writer isn’t committing to the certainty that periods communicate, and is unsure of how to continue, creating an shaky tone. One surveyed user characterized the use of comma ellipses as “something along the lines of wanting to say more, but [not being able to] find the right words to say or type,” and another characterized the use of comma ellipses as “mak[ing] a sentence feel like it’s hovering.” Another user commented that the obvious misuse of commas with regards to formal english grammar rules also contributes to the unsure tone the sentence is given—“[the commas] aren’t being used normally, and their being out of place shows how strange or overwhelming the situation is to the subject.” This uneasy tone is further amplified by the fact that traditional period ellipses do not usually seem to carry any specific connotation, but are rather a somewhat emotionless, straightforward trailing off. To readers, since comma ellipses deviate from the norm, they must represent something more than what normal ellipses represent.

        The fact that traditional period ellipses do not seem to carry any specific emotional connotation as well also contributes to the fact that comma ellipses can also be used to make a statement seem less serious. Period ellipses come off as much more formal and structured to begin with because they are an existing, acknowledged convention in the English language, so the fact that they carry no emotional tone only adds on to the formality of the ellipses. By this logic, comma ellipses, which are not a standard convention in the English language, seem more relaxed in comparison, since they are not “correct” by any formal standards. Thus, the commas come off as more naturally human, and, as one user wrote, more genuine. More interestingly, one user in particular noted that when they use comma ellipses, they see the “,,,” as “sweat drops, to express hesitation,” thus revealing how to some, the commas literally become representative of something that is human nature. It is no secret that when things become more natural and casual, people begin to relax and feel more comfortable as the seriousness of the situation has decreased, and the use of comma ellipses is no exception to this phenomenon.


        As I mentioned before, I only analyzed five different linguistic phenomena that are commonly used on Twitter—far from the total number of existing linguistic phenomena that are commonly used on Twitter, and extremely far from the total number of existing linguistic phenomena that are commonly used online and on social media in general. However, I still believe that the analysis that I did complete is indicative of the many wonders that online slang and conventions achieve linguistically, such as changing the meaning of a sentence with a single character and finding intuitive ways around the roadblocks that reading text presents to human communication. Really, some may argue that online venacular is becoming its own language, which I am beginning to agree with—my parents do not understand a lot of the online colloquialisms that I use in messages to them, and I remember keyboard smashing to a friend my age who does not use Twitter in lowercase letters as a filler and receiving an extremely confused reply. Perhaps spending time on social media was not a waste, but rather, a time for me to practice a “foreign” language.

        Another thing that I would like to note in my analysis is simply how complicated many of these linguistic phenomena are. An extremely large proportion of the responses in my survey in explanation boxes contained responses along the lines of “I can’t really explain why… it just makes sense,” which is extremely true—it is hard to explain why some letters look soft, or why a comma looks less certain than a period, and there might not be a definitive explanation at all to why certain slang works and why others do not. My point is, I do not know whether it is because I am a teenager who feels far away from adulthood, but I feel as if many professionals and adults tend to overlook linguistic phenomena that are birthed from the internet because they see it as just a blip in existence that exists because of children playing around online compared to the many centuries that the formal English language, a language used by great scholars and philosophers and academics, has existed for. From my point of view, online linguistic phenomena are not simple blips. As seen from this survey, they are extremely complex signified meanings masked by extremely simple and elementary signifiers, and thus, should not be overlooked at all. Technology and the internet will only grow and become more prevalent as time goes on, and colloquialisms like the ones I analyzed in my survey and in this paper will become commonalities among all.

Simply put, all I hope for is that my linguistic analysis shed some light, even if just a single ray, on the sheer complexities of online language. Like many established languages, it is a continually evolving tool for communication with quirks and systematics that can be analyzed with just as much meaning as formal English can.