hpr-knowledge-base/hpr_transcripts/hpr0154.txt

Episode: 154
Title: HPR0154: Linguistic Public Radio Episode 0
Source: https://hub.hackerpublicradio.org/ccdn.php?filename=/eps/hpr0154/hpr0154.mp3
Transcribed: 2025-10-07 12:28:59

---

Welcome to another episode of HPR.
In today's episode we will be hearing audio from linguistic public radio, episode 0.
linguistic public radio is a podcast for linguists and people who like languages, you should
check them out on linguistic chat.org, they have an IRC channel, a forum, and a little
community that just launched as of today.
Their podcast will be airing every Monday on their site, and we will be airing some of
their shows here on HPR.
Welcome to the first episode of linguistic public radio, the radio associated with language
chat.org.
I am Plexi and I will be your host this evening.
This show will have a weekly Monday rotation with extra shows on Thursdays, we are still
recruiting podcasters to go on our rotation, so if you are interested, please email me.
There is going to be a few series, there is an Esperanto tutorial series, a series on
typology of your languages, a series on language acquisition, a series on computational linguistics
as well as other shows that are random about whatever the host wants to talk about.
My first episode is about language learning and transfer.
To those who don't know what transfer means, it's a term coined for second language learning
when the learner's native language interferes with the production of the second language.
There is positive and negative transfer as well as conscious and unconscious transfer.
Positive transfer is when the two languages are alike.
For instance, English has something in you learning German that has the same thing and
you transfer a good rule from English to German.
Negative transfer is when the two languages are different in something and the learner mistakenly
transfer is wrong information to the target language.
Let's call the first language L1 and the second language L2 for now.
There are several theories on transfer.
Some people believe the initial state of the second language is the same as the final
state of the first language, which means that at the beginning of learning a language,
you transfer absolutely everything.
Basically, you're just inputting, you're just plugging in the words of the foreign language
and the grammar of your own language and then a little by little you start learning to
not transfer and you only take out the bad rules and you keep the good rules.
This is called total transfer.
Other people believe that you only transfer plausibly similar things, so you don't assume much
at the beginning but then when you only transfer things that seem to work fine,
you know, so partial transfer.
Some people believe that you don't transfer anything at all and that when learning in
your language, the individual starts with a blank state with no transfer at all, which
is called the no transfer theory, of course.
Surprisingly, some people actually believe that even though there's wide proof that a lot
of like a lot of language learners transfer.
Anyway, there's a theory called the developmentally moderated transfer hypothesis and it goes
something like this.
The parser for the second language is assumed to be distinct from the parser for the first
language.
Therefore, it cannot process complex structures in the second language, even if there are
identical structures that occur in the first language.
One example for this is given by Haconson, I don't know if I pronounced this correctly,
it's the link and the show notes.
About native speakers are Swedish who are learning English.
They focus on a complex structure, it's the fact that both German and Swedish have a
subject verb inversion when the adverb is funded, so when the adverb is at the beginning
of the sentence, they switch the verb and the subject.
If Swedish learners of German did switch the subject verb and German, that would be a case
of positive transfer.
However, they noticed that the Swedish learners do not produce German sentences correctly
and they do not do the inversion from which they conclude that the positive transfer did
not happen, which means that even though the learner's first language parser is capable
of parsing verb-inver-subject inversion, the second language parser is not.
I have no idea why this hypothesis is called developed, developmentally moderated transfers
since it says that transfer only occurs when the second language parser is ready, which
means that there's no reason this is transferred at all because we can't tell if it only
occurs when it's ready, then it's just, you know, learning just regular old language
acquisition.
Frankly, in my opinion, I think that transfer relies on an assumption that it's the same
parser or at least a parser that interacts with other parser at least, and it's just
learning new tricks.
So here's the interesting thing about the study that I just quoted.
So Swedish students that are learning German had already taken English courses.
Why is this interesting?
Because what if they were transferring incorrectly from English that inversion does not take
place?
Okay, imagine the situation, you're a native speaker of English, English sentences normally
start with a subject.
You take three years of Spanish in high school, Spanish sentences often omit the subject
and start with the verb immediately.
A few years later, you start taking French classes and you're trying to speak French in
this class and your brain goes into foreign language mode.
Your parser then brings up all the rules or whatever that are labeled foreign language
rules or whatever.
And it applies them to French, which results in incorrectly producing French sentences
without the subject, even though you're fully capable of producing sentences with subject
and you do it every day in your own language.
And you would be capable of producing them in French, except you just don't know.
In other words, the second language would be blocking the first language from providing
positive transfer to the third language.
Anyway, one thing to note about transfer is that it only occurs in learning and it does
not occur in the context of language acquisition in children.
In other words, if your mom is French and your dad is Swedish, you should be able to acquire
both French and Swedish independently with no transfer between the two languages, which
is really interesting because it shows that the processing of language happens drastically
differently between children and adults.
That's all for today.
This was Black Scene.
Thank you for listening to H.P.R. sponsored by Carol.net, so head on over to C-A-R-O-DOT-A-C-R-R-R-R-O-DOT-A-C-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R-R
R-R-R-R-R-R-R-R-R
R-R-R-R-R-R-R-R