Idiolect - Do You Have One?
The word "idiolect" is formed from combining a back-formation of the "-lect" part of "dialect" and the Greek prefix idio- (meaning "own, personal, distinct, peculiar"). So an idiolect is the distinctive and unique way a person uses language. It's also a debated topic within linguistics, especially about whether they exist and/or can be described.
I've mentioned before, when describing linguistic profiling, that some of a person's demographic info (gender, age, education, etc.) can potentially be deduced from their language use. It's too easy to get philosophical about individuality and idiolects. Look into social statistics to get a sense of what I mean by that. Let's instead keep the focus on you. Let's talk about your idiolect.
Like many aspects of your identity, the way you use language is shaped by the kind of human you are and the life you've lived. There are so many factors of your life that have shaped the way you communicate with the world. Your voice is determined partly by your sex and the shape of your throat and face, for example. Your spoken language, if you do speak, is influenced by where you learned your first language and who you learned it from, and how. Your written language reflects your education especially, such as where and when you learned to write. These are just a few examples of the way your life influences your language.
While we consider all of these features of life and language, we can't forget that life is a process. As you're a living person, your life goes on, and it will continue to influence your language. You, and everyone else, can also change the way you use language. Learning more languages, for example, or incorporating new slang into your vocabulary. On top of all of these fixed or mercurial aspects, there are factors of influence that are temporary - even ephemeral. Your mood, health, sobriety, and so on, also affect your language.
This is all to say that, were you to attempt to define your idiolect, you couldn't use just the span of a single day. Ideally, you would want to build a corpus of your own language samples from a span of years at least. Seems like an impossible task, right? It probably is. That is why idiolects don't really fit into forensic linguistics too often. If building a corpus of your own language is difficult, imagine trying to build one for an uncooperative or unknown suspect.
There is a process similar to building an idiolect corpus in authorship cases, though. Authorship attribution or authorship analysis cases are those where a forensic linguist compares known samples from a suspected author to determine if they likely created the questioned sample. Language samples known to be created by the suspected author are collected for comparison, as many as possible. Ideally, those samples come from the same register, or communication context, as the questioned sample. For example, a questioned email is best compared against other emails, rather than text messages or love letters. Luckily for forensic linguistic analyses, though, completely comprehensive idiolect analysis is not absolutely necessary. However, when it comes to language data, more is better.
I did analyze an idiolect once. Technically, the complete idiolects of two individuals. Fictional individuals: Gollum and Sméagol. Since they only use language in a few books, and die in one of those books (spoiler alert), and the author is deceased, it was possible to collect their every word. I analyzed their idiolects and wrote a paper on my findings. If I ever get copyright permission from Harper Collins, I will publish and share that paper. Still, it is an example of what it would take to completely document an idiolect.
I want to conclude by saying that I do believe that idiolects exist, but they're nearly impossible to completely document and define for real individuals. However, the concept of idiolects is, to date, theoretical. As a theory, conceptual idiolects do have value in linguistic analysis.
I've mentioned before, when describing linguistic profiling, that some of a person's demographic info (gender, age, education, etc.) can potentially be deduced from their language use. It's too easy to get philosophical about individuality and idiolects. Look into social statistics to get a sense of what I mean by that. Let's instead keep the focus on you. Let's talk about your idiolect.
Like many aspects of your identity, the way you use language is shaped by the kind of human you are and the life you've lived. There are so many factors of your life that have shaped the way you communicate with the world. Your voice is determined partly by your sex and the shape of your throat and face, for example. Your spoken language, if you do speak, is influenced by where you learned your first language and who you learned it from, and how. Your written language reflects your education especially, such as where and when you learned to write. These are just a few examples of the way your life influences your language.
While we consider all of these features of life and language, we can't forget that life is a process. As you're a living person, your life goes on, and it will continue to influence your language. You, and everyone else, can also change the way you use language. Learning more languages, for example, or incorporating new slang into your vocabulary. On top of all of these fixed or mercurial aspects, there are factors of influence that are temporary - even ephemeral. Your mood, health, sobriety, and so on, also affect your language.
This is all to say that, were you to attempt to define your idiolect, you couldn't use just the span of a single day. Ideally, you would want to build a corpus of your own language samples from a span of years at least. Seems like an impossible task, right? It probably is. That is why idiolects don't really fit into forensic linguistics too often. If building a corpus of your own language is difficult, imagine trying to build one for an uncooperative or unknown suspect.
There is a process similar to building an idiolect corpus in authorship cases, though. Authorship attribution or authorship analysis cases are those where a forensic linguist compares known samples from a suspected author to determine if they likely created the questioned sample. Language samples known to be created by the suspected author are collected for comparison, as many as possible. Ideally, those samples come from the same register, or communication context, as the questioned sample. For example, a questioned email is best compared against other emails, rather than text messages or love letters. Luckily for forensic linguistic analyses, though, completely comprehensive idiolect analysis is not absolutely necessary. However, when it comes to language data, more is better.
I did analyze an idiolect once. Technically, the complete idiolects of two individuals. Fictional individuals: Gollum and Sméagol. Since they only use language in a few books, and die in one of those books (spoiler alert), and the author is deceased, it was possible to collect their every word. I analyzed their idiolects and wrote a paper on my findings. If I ever get copyright permission from Harper Collins, I will publish and share that paper. Still, it is an example of what it would take to completely document an idiolect.
I want to conclude by saying that I do believe that idiolects exist, but they're nearly impossible to completely document and define for real individuals. However, the concept of idiolects is, to date, theoretical. As a theory, conceptual idiolects do have value in linguistic analysis.
Comments
Post a Comment