On Thu, Jan 14, 2016 at 10:20 PM, Nicole Valicia Thompson-Andrews
<[log in to unmask]> wrote:
> Hi, what are some ways alien can acquire languages?

We don't know because we've never met any aliens.  But I suppose the
basic two ways humans acquire languages are immersion and formal study
-- usually, with second languages, a combination of the two.

Here are some excerpts from my work in progress on conlangers who are
fluent in their conlangs:


Implications from second-language acquisition research

How does one acquire fluency in a language that no one else speaks?
Stephen D. Krashen, a language acquisition linguist, writes:

".....The Comprehension Hypothesis claims that language acquisition
does not happen when we learn and practice grammar rules. Language
acquisition only happens when we understand messages."

How if at all does this apply to a lone conlanger learning a language
that barely exists yet, and which will probably have no other
speakers? Clearly in order for the conlanger to "understand messages"
written or spoken in their conlang, they must first create those
messages; and to create the first such messages they have to
consciously apply grammar rules. But it's not necessary, as most of us
know by experience, to have all the grammar loaded into skill memory,
or to thoroughly know all the words, in order to start creating text
in the conlang, whether original or translated. This theory of
language acquisition suggests that we are probably better off not
trying to memorize the grammar rules of our conlangs per se, but
rather to use them in creating text, and then let the rules soak in
unconsciously while we re-read the texts many times. It also suggests
that the themes we choose for original writing or translation into our
conlangs should be matter that's interesting enough we won't mind
re-reading it over and over. That may seem too obvious to need saying;
but it does suggest that writing connected continuous text — whether
stories or diary entries or what have you — is more likely to help us
acquire our conlangs than just writing example sentences illustrating
usage of particular words and various points of grammar; and also,
that it's the re-reading of these texts, rather than the writing of
them, that helps most in learning the language.

For acquiring the spoken forms of our conlangs, this theory of
language acquisition would suggest listening to recorded speech over
and over; but to listen to such recordings, we need to first create
them, which seems to imply that we need to learn to speak the conlang
before we can learn to speak it. The fact that a few conlangers with
no one else to speak their conlang with have actually learned its
spoken form with a reasonable degree of fluency suggests that this
paradox is not, in fact, as binding as it first appears. For one
thing, it's possible to rehearse reading from a written text several
times before recording it, and then, perhaps, to edit out any
remaining disfluencies with sound editing software; thus one can get
clean recordings of a hypothetical version of oneself who is more
fluent in the conlang than one actually is as yet. I suspect, however,
given the answers people gave to my question about listening
comprehension of recordings they've made, and to the question about
people's methods for learning their conlang, that most conlangers who
have become fluent in the spoken form of their conlangs did so
primarily by talking to themselves aloud, at first hesitantly and
disfluently, and gradually more confidently and fluently over time.

It also seems possible that this language acquisition theory describes
an effective method for the majority of people, but that this method
may not be the most effective for a minority with different learning
styles; perhaps conlangers are likely to be in this minority for whom
conscious study of grammar is relatively more effective than for most

John Schumann's acculturation hypothesis says that fluent
second-language acquisition tends to happen as a by-product of
successful acculturation, that is, social and psychological adjustment
to an adopted culture. He was apparently focusing primarily on
immigrants who are trying to adapt to the culture of their adopted
country and acquire its language at the same time. Can we apply this
to the learning of conlangs? Are people whose conlang is associated
with a culture within a well-developed fictional world, and who
strongly identify with this culture, more likely to become fluent in
their conlangs than those whose language has no imagined culture
attached? Is acculturation a causal factor for fluency in itself, or
only by way of giving people strong motivation to use the target
language more often?

Elizabeth Barrette wrote in response to my survey:

When I make a concerted effort to learn a language ..... [I] [c]hoose
a name to be the part of myself who will be a native speaker in this

Tony Harris wrote:

I think it was more how I immersed myself in the culture and
"heritage" of the Alurhsa people that caused me to immerse myself in
the language as well. ..... I have come to identify "spiritually" in
some way with the Alurhsa world-view [or] ethos.

These quotes suggest that strong identification with the conlang's
associated culture may be helpful for some conlangers, but I can't yet
draw any strong conclusions about this from the survey response data.
Comparing the more and less fluent groups of conlangers, I don't see
any strong correlation with whether their conlang is associated with
an imagined world or culture; if there is a correlation, it seems to
be negative but weak. However, I plan to try to find out more when I
do my follow-up survey. I will need to ask more specific questions
about how thoroughly developed people's imaginary worlds are and how
strongly they identify with them, and control for several other
variables, because I suspect that conlangs associated with an imagined
world are liable to be more difficult on average than those that are
not; making a language naturalistically plausible and and making it
easy to learn are more or less contradictory goals. I started desiging
gjâ-zym-byn as a personal language with no associated fictional
culture partly because the more or less naturalistic artlangs I had
been working on were too irregular to be easy to learn, and it seems
that I'm not alone in this; at least five of the twelve respondents
who answered yes to the question "Can you speak spontaneously in your
conlang at conversational speed?" have auxlangs, engelangs or
non-naturalistic artlangs with an unusual degree of regularity or

Is it possible to compare the results of my conlang fluency study, or
of any future study along these lines, with previous studies of how
people learn or acquire natural languages? The problems with comparing
the results of these studies are severe. Most natural language
acquisition studies focus on a group of people learning the same
language — ideally, all with the same native language, or all learning
the target language in the same place at the same time — and make
observations on the order in which they tend to learn the target
language's various features, what features give them most trouble, and
what factors (demographic, psychological, etc.) correlate strongly
with the learners' rate of learning and degree of success. A conlang
fluency study has to deal not only with diversity of learners, but
extreme diversity in the target languages and in the conditions under
which the learners are studying them; further, the researcher is
unlikely to be knowledgeable about most of the languages involved in
the study, and thus has to rely on self-reporting for data on how well
the respondents have mastered particular features of their conlang.
There are whole subfields of second-language acquisition study whose
findings are impossible to apply if the researcher doesn't know the
target language being studied. In short, I doubt I will be able to
compare results from this study directly with results of natlang
acquisition studies; but I hope to come a little closer with a
follow-up study, consisting of additional in-depth questions sent to
the most fluent respondents to my earlier survey.

Jim Henry