Four explanations for the nature of names
Guest Article by Mr. Sten Lillieström, Next Venture (Next Venture acquires taken domains, or helps you find a new name identity with the domain available for immediate purchase; LinkedIn)
For the duration of the existence of domain registrant rights, a common argument for good faith in the face of existing trademark rights has been whether the contested string is a dictionary entry. Is this an appropriate measure for a domain arbitration panel or court?
Even though no mention of the dictionary word “dictionary” is present in any guiding statute related to trademarks and domains, it is nevertheless everpresent in actual practice related to cybersquatting. UDRP decisions for example are littered with the term, and it is commonly invoked for much of the same reason. (Ironically on both sides of the aisle, albeit with different aims.)
Complainant counsel argue that since the trademark of their client is not a “dictionary word” – the Respondent is more likely to be cybersquatting. Respondent counsel on the other hand argue that since the domain name of their client is a “dictionary word” – the Complainant’s claim must be inherently weak. This is such a common argument that it’s “part of the furniture”, and perhaps for this reason difficult to consciously recognize and evaluate.
If something is in the dictionary, so the argument goes, this corroborates the claim that no targeting of a specific trademark was intended. It’s apparently a common word, accessible to anyone. It’s therefore deemed more feasible that the registrant simply picked it out of the public domain for the reason that it is the common word available to them, and with the likely intent to use it in that sense.
The problem with this interpretation is that it does not even start to account for actual circumstance. (One place to start for the blissfully oblivious is this simple and instructive talk by Linguistics Professor Anne Curzan on what a “word” is.)
Sure, brand names sometimes are “dictionary words”, or consist of them, but they are virtually never used in their prescribed literal sense. Even in the rare case that a brand name consists of a dictionary word, it is with few exceptions employed with a pinch or even a pound of creative spin, intended to create a distinctive identity. The context of the usage and the terms selected are almost never intended to be merely literal, but a figurative and faceted statement that results in an identity that transcends mere literal description. This is consequently built into the nature of the demand for prospective brand names, and arguably the main reason that a moneyed market for domain names exists.
It is therefore an astonishing act of intellectual acrobatics to on one hand acknowledge that investing in domain names with the intent to resell them is perfectly alright, but at the same time disallow the generic mechanisms that create appeal in names.
In general however, and to further complicate things, brand names are not dictionary words at all. They are intended to be novel and to employ meaning and aesthetics in ways that go beyond whatever words may be recorded in dictionaries.
This does however not mean that they are particularly “fanciful” by the standards of naming, or actually even by the standards of natural language at large.
Here are the main reasons why:
Morphology.
What we call words have constituents. These building blocks are called “morphemes” and are commonly utilized both in deciphering the meaning of a word construct partly new to us, and in creating novel words. The concept may be modern by historical standards (1880’s) but is really quite established by contemporary standards, and should not be possible to ignore without deliberate effort. In UDRP decisions however, the term is virtually non-existent.
Morphemes are not listed in dictionaries, but they are nevertheless a pervasive and universally accessible feature of language, and consequently also a cornerstone in the field of Linguistics. See for instance Words and Rules by Steven Pinker for a comprehensive exposé of what goes on under the hood in terms of Morphology when we use language.
In the context of brand name creation, morphemes are naturally ubiquitous to the extent that denying their existence and relevance can simply not be interpreted as anything but a display of willful ignorance.
In natural language, morphemes can be one letter/sound long and there is really no limit to how long they can be. They are the smallest units of language that have lexical meaning. So for example, “crocodile” is a morpheme, but so is “-s”, if added at the end of a noun. Suffixes such as “-ance” or “-ly” or “-ment” are other examples, instrumental in allowing us to alter word classes. There are prefixes and circumfixes and infixes as well, that make it possible for us to express ourselves more distinctively.
This knowledge is universally accessible to anyone familiar with English. A crushing majority of languages use similar devices to alter and construct meaning. It is as a consequence also naturally prevalent within the context of naming and brand naming.
The well known European security company “Verisure” for example is a meaningful combination of morphemes. The morpheme “ver-” or “veri-” as seen in words such as “verification” or “veritable” is of Latin origin and means “true”. It’s combined with “-sure”, which in addition to being a word/morpheme on it’s own, is also a common suffix that may for instance transform verbs into nouns, a process known as nominalization. Something very suitable in brand name context. Two examples out of many in natural language are the nouns “composure” and “exposure”.
Another similar example in the brand name sphere is “Accenture”.
So even though “Verisure” or “Accenture” does not have a matching dictionary entry – they build on linguistic principles available to all, and evidently also employ lexical meaning directly related to the nature of the use case. It is therefore natural to atleast entertain the idea that even if “Verisure” did not happen to ever coin “Verisure” – that someone else eventually would have. And perhaps even that someone not cognizant of the definitive existence of one or several “Verisure” would coin it independently, for similar reasons.
Although this is certainly not peripheral or exotic knowledge, UDRP practice sometimes seems to go to great lengths to avoid addressing it.
In the decision <Finwise BV v. Anne-Jan Hempenius, NameAvenue, Case No. D2020-3135> for instance, the Panel concludes that since “Finwise” is not a “dictionary word”, it is by design not considered a bona fide domain registration for a domain name investor. “The difficulty for the Respondent is that “finwise” is not an acronym, dictionary word or common phrase, as far as the Panel is aware… …The Panel is unable to distil any generic or dictionary meaning of “finwise” from the variety of contexts in which it is used. Accordingly, the Respondent has failed to provide evidence that the disputed domain name is a common, dictionary term that he acquired for resale.”
At the same time, a linguistic analysis of the terms comprising the domain name would inarguably establish that this conclusion is highly questionable. One might even argue that it is intellectually dishonest in the face of readily available knowledge.
Phonology.
Language consists of sound that is arbitrarily paired with meaning. There are relevant exceptions, but this is largely the case. There is for example no particular reason that the word for dog is “dog”, specifically.
Different languages however employ sound patterns unique to them, which is why when learning a new language, that we tend to inadvertently color it with the sound patterns of our native tongue. These sound patterns are largely inherited conventions, but also delimited by the capacity of the speech organs, as well as the economy required in natural flow of speech.
Accordingly, even though the pairing between sound and meaning is largely arbitrary, it is certainly not random at the systemic level, and quite available for prediction. There is even a subfield to Phonology called “Phonotactics” which employs statistical likelihood analysis on corpus data of sounds and sound combinations. Anyone attempting to select or assess a brand name will most likely make judgments that subconsciously consider Phonotactics. For further reading on Phonotactics, the paper Phonotactic Probability of Brand Names: I’d buy that! may provide additional insight.
The majority of brand names employ phonological assessments in that pronounceability is a key aspect. This is not really a choice we make. A good name is easy to say by both design and definition. It is something we naturally gravitate towards – if we can manage to not overthink things.
Virtually any prominent brand name example is relatively easy to say. “Coca-Cola” could not have equally been “Krilgo-Lysne”. “Xerox” could not just as well have been “Skzreog”.
Adherence to phonological principles of English excludes the vast majority of possible brand names and therefore increases the likelihood that brand name selection efforts may independently overlap; increasingly so since brand names invariably also employ brevity as a key aim.
Syntax.
Language is not a bowl of word salad. It has structure and logic that scientists have spent much time and effort excavating and formalizing. Words and morphemes for example are strung together based on a system that may be difficult to discern, but also evident once formulated. Like any subject matter in the study of language, Syntax is not a choice. We employ it like we employ breathing, unconscious of the mechanisms involved. Remove it however, and language falls apart at the seams.
We may think that grammar and syntax surely has no relevance in the context of brand names, but that would be wrong.
Any utterance or expression is equally governed by these rules, which by the way should not be confused with the conventional concept of “correct grammar” you may have learned in school. Syntax is a scientific concept that describes the combinatorial nature of language, and not a prescriptive set of table manners.
A quintessential example of the delineation between syntax and meaning (semantics; below) is a sentence contrived by renowned linguist Noan Chomsky which shows that even though there is no obvious literal sense to make of a sentence, that it can nevertheless be perfectly grammatical. “Colorless green ideas sleep furiously.” (Seminal work by Chomsky can be found in Knowledge of Language: Its Nature, Origin, and Use. A fascinating account of the minimal mechanism required for Language to exist can be found in the more recent work Merge and the Strong Minimalist Thesis.)
Brand names in general inherently adhere to Syntax. Sometimes they diverge from how something would be put in natural language, but the reasons for this are the unique conventions governing brand name composition, and not an exception to Syntax writ large.
“Facebook” for instance is a closed compound word form that is grammatical in several ways, even though it is currently conventionally off limit to compound this way, without a space, within conventional English orthography. “Bookface” would not have been an equally viable option, for syntactic reasons, and considering Syntax, “FacesBook” would have been an ungrammatical and thus non-existent option.
The addition of Syntax even further decimates available options and increases the likelihood that name selections may independently overlap, irrespective of whether some specimen may have been selected to feature in a dictionary.
Semantics.
The study of meaning as it relates to language is called Semantics. If Syntax describes the rules for stringing words together into sentences, Semantics is the theory of what the words mean and how. Semantics are a tricky beast since the boundaries between language and other features of the mind may tend to blur. For more than a handful on this subject, the The Linguistics Wars by Randy Allen Harris is a good read. It may however be clear enough that linguistic expressions prompt interpretations of meaning and that there are universal mechanisms for humans to achieve them.
Language works astonishlingy close to telepathy in that we speak our minds in an effortless flow, and the contents of our minds are instantly transferred into that of another human being. Semantics and whatever drives it must be a big part of the explanation for this magic trick.
Recent studies accordingly emphasize that analogy and metaphor are not some artful exception or odd literary device, but deeply embedded in the way we process thought. (See for instance The Way We Think: Conceptual Blending And The Mind’s Hidden Complexities by Gilles Fauconnier and Mark Turner). Similar studies may imply that emphasis on literal meaning and/or neglect of figurative meaning within trademark practice is not in agreement with how the human mind actually works, something that is generally regarded as a fundamental basis as well as a goal for the application of trademark law.
Brand names are rarely literal, and commonly figurative. This is widely acknowledged by corpus studies of actual naming specimens, and it should come as no surprise that the very same devices are freely and legitimately employed by people at large in general, and by naming professionals in particular.
The existence and use of Semantics naturally even further decimate the viable options available for name identity selection.
I’m additionally adding two relevant but perhaps more intricate learnings along the same lines as the collected quad-unity of Linguistics; Morphology, Phonology, Syntax and Semantics, above, just to underline the complexity of the matter at hand.
(Sound Symbolism.)
Contemporary studies show that something called sound symbolism may affect linguistic preferences and perhaps particularly naming devices a lot more profoundly than commonly thought.
Even astute academics in the field of Law find it relevant to address this issue. For further reading, one example is a paper entitled Are Trademarks Ever Fanciful? by Loula Fuller & Dan Myers Professor Jake Linford at Florida University.
Simply put, sound symbolism is the exception to the rule that sounds in language are arbitrarily paired with meaning, and it may be a quite comprehensive exception. This exception is however completely unaccounted for within the body of trademark law.
One example are units called “phonesthemes” that are language specific and can be used to group words together based on the shared relation between sound and meaning. “Gl-” for instance is a statistically significant marker for meaning related to shine and sparkle. “Glitter”, “glisten”, “glow”, “gleam”, “glare”, “glint”, “glimmer”, “gloss”, the list goes on.
Other more universal aspects of sound symbolism include singular phonemes at large. One of the first studies on the nature of the relation between sound and meaning is called the mil/mal experiment, conducted in 1928 by anthropologist and linguist Edward Sapir. (Most known for the contested Sapir-Whorf hypothesis that vocabulary dictates how we see the world.)
Similar studies have followed, such as the kiki/bouba effect and the takete/maluma phenomenon. A recent study dubbed A Sound Idea: Phonetic Effects of Brand Names on Consumer Judgments explains how the manner in which phonetic effects of brand names manifest in consumers is automatic, uncontrollable, outside awareness and effortless.
Needless to say, instilling preferrable meaning through knowledge about sound symbolism can be invaluable to brands. The option further decimates available naming options for those who wish to employ these scientific results in their naming process.
It may at this point have become more difficult to argue that since something is not in the dictionary, it is not a legitimate investment option. Our linguistic and onomastic knowledge and natural tendencies make viable naming options, irrespective of dictionary inclusion, both publicly available and generally valuable. A growing market for brand type naming options in the form of naming agencies and domain marketplaces further underline this fact.
Below is one final example of how language and therefore naming actually works, this time, loe and behold, it consists of a principle that is not exclusive to language and that furthermore is quantitative. Something that every engineer can appreciate.
(Statistics and the Tolerance Principle.)
Traditionally, as far as modern Linguistics go, many theories and models have tried to make Universal Grammar manage way too heavy lifting. This is what linguist Charles Yang thinks in his recent work The price of linguistic creativity: how children learn to break the rules of language. (Universal Grammar is the theory that our language capacity is genetically endowed and that we develop a language primarily from an initial state.)
According to Yang, and his research and results have met wide acclaim and corroboration, there is a principle that governs when a linguistic rule activates and de-activates, and it’s purely a matter of numbers. It’s so completely a matter of numbers that it can be formulated into a mathematical formula.
Simply put, the productivity of a rule that for example says that the past participle of verbs is generally “-ed” depends on the relative number of exceptions to this rule. We may surmise that since we have encountered for example 18 examples of “-ed” that this is a rule that it is productive to extend to a general rule. We can therefore “know”, with very scarce evidence, that it is sufficiently correct to employ “-ed” for any verb that does not have a memorized exception, based on experience alone.
As far as brand names go, this may imply that certain forms can generalize and also suggest how. We may periodically see examples of similar things at play when a certain way to name catches on. Spotify started a “-ify” trend. Another example may have been the “-ly” trend, yet another the cutting out vowels trend (known as “disemvoweling”).
These trends will rise and fall and come and go, perhaps as a cosequence of something like the Tolerance Principle. They are therefore certainly not inherently “bad” or “incorrect” or anything of the sort. Admittedly, when the aim is to be unique like it is with brand names, it does not help if you use a naming mechanism that is already virtually a rule, but in general, this regulates itself just fine.
It is however prudent to note that a domain investor may be able to quantitatively and accurately predict whether a certain naming pattern will catch on (to eventually self-destruct, and sometimes to rise from the dead.)
There may be other applications of the Tolerance Principle yet to be discovered. In any way, it may be viewed as yet another mechanism by which naming can be predicted, and that has nothing to do with dictionary inclusion and everything to do with the way we actually employ language.
Possible effects.
Even though these established learnings about language are quite common knowledge, they rarely manage to pierce the veil of conventional wisdom. It could in fact be argued that in the contexts where they are of use, that they are being willfully ignored. The reason for this may be difficult to speculate about, but a wild guess may be that status quo is preferred over change, however necessary change may be to achieve improved legitimacy and functionality.
It is in any case inavoidable that even the general proliferation of status quo has consequences in itself.
One of them may be UDRP decisions that struggle to properly address the possibility that a name could have been independently coined and acquired for reasons unrelated to some questionable trademark right. Other effects may be misguided counsel from trademark attorneys or brand protection professionals that may sometimes be inclined to believe that since something is not the literal dictionary word, that their client is being targeted even though that may not be the case.
These effects may indirectly harm the development of a solution to demand that is generally preferred in modern society, namely the formation of a functional and legitimized market.
-Sten Lillieström-
As the founder and CEO of Next Venture*, a domain acquisition brokerage and naming firm, I help entrepreneurs and businesses find and secure their desired brand names on the internet. As a consequence I also operate a portfolio of domain names that serve as viable naming options in different niches and industries.
This is part of a series of occasional guest articles. The opinions of the author are their own and are not necessarily reflective of the ICA’s.