11 thoughts on “Comments to the phonological acquisition’s paper”
I found this article very interesting. I had heard in other classes that acquisition of L2 phonemes that are similar to L1 phonemes was more difficult in terms of producing a native-sounding segment, but not much more detail was given and perception was not addressed. It sounded somewhat similar to the Speech Learning Model she mentioned. It is vague though as far as what is “similar” and “new”. The experiments that she did to how that it is the features of the phonemes, not the phonemes themselves, that make the difference in whether they can be acquired were fairly convincing. Though I would imagine that it would make a difference if the phonemes of one language are in an allophonic relationship in the other or simply not present. She mentioned that l/r are both a flap in Japanese so the fact that they are perceived categorically is not a surprise. The evidence from the other L1s she tested still makes me think she is right that features not present in the L1 will hinder acquisition of phonemes in the L2 that depend on that feature, and L1 features that are present will help acquisition of L2 phonemes that rely on those same features. It just seems it would make it that much more difficult if sounds that in one language are phonemic are allophonic in another, compared to sounds that are phonemic in one language but entirely absent in the other.
The most interesting aspect of this article for me was that it discusses that L2 phoneme acquisition is affected not only by language acquisition, but phonological representation, and speech perception as well.
I have never thought about much about speech perception affecting speech production before. Of course logically it makes sense because if you can’t hear a difference between two sounds, how are you supposed to produce that difference?
The article says that L2 learners try and categorize phonemes based upon the ones they already recognize in their L1. And there is an order that they categorize them in based upon certain features. It also says that as infants we have a greater ability to distinguish contrasting sounds in all languages but as we get older this ability begins to degrade from all potential contrasts to only native ones. This is explained by the fact that when we filter out only the contrasts that create different meanings in our native language, it makes processing much faster.
For me thinking about this raised the question of how easy is it for L2 learners to start listening for contrastive sounds again after they are exposed to a new language?
It also makes me think that in a way perhaps when learning a new language it is good to take a phonetics course in that language to try and at least become aware of the sounds that are produced differently in that language so that you can better try and perceive the differences.
One more thing that interested me is that I have a professor who is partially deaf, but she can speak at least 3 languages fluently. This makes me wonder if speech perception really matters all that much because someone who’s ability to perceive speech is impaired, but is still very successful at producing the different phonemes of different languages.
As she says in this article, I can say it is true that for Japanese it is hard to distinguish some English sounds such as l and r, or s and th. When we describe English words contain sounds of th and s such as “gas” and “path”, we use the same character for “s” and “th” part because we don’t have way to express the sound perfectly. So, we almost have no chance to know that there is a difference between the sound “s” and “th” and it leads our lack of ability to distinguish them. Also, most of Japanese learn English by using Japanese in school, so that they may naturally acquire English based upon their first language, Japanese. I wonder that if we learn L2 by using only the language, our phonological acquisition can be better or not.
Two interesting and correct elements that shape the phonological acquisition for non-native speakers; the factor that determines the phonological acquisition is the L1 features and not the segments, and the other fact that speakers’ ability to acquire novel contrasts which do not exist in their L1 grammar does not change over time or by developing their L2. I applied these elements on the Turkish language speakers who are learning or speaking English language. Turks are unable to produce the sounds /w/, /θ/, and /ð/. Same situation applies when producing /b/ instead of /p/ by the Arabic language speakers, due to the fact that these contrasts do not exist in their native L1. But accuracy in performance was different, as the Turkish speakers are unable to produce these contrasts regardless of the number of years of exposure to English language, but Arabic speakers were significantly developed their performance over time.
While reading this article, I was reminded of the exercise we did in class, trying to distinguish /p^h/, /b/, and /p/ sounds, as well as d, t, and th sounds (I can’t remember what language we listened to…). With that exercise, we talked about our cognitive “categorization skills” determine how we differentiate these sounds. In this article, this categorical perception is pointed out to perhaps be a hindrance to second language acquisition :
Yet although categorical perception aids in processing of one’s native language, it can be a barrier to correctly perceiving and processing a foreign language: variation in the acoustic signal which is filtered out by the native phonological system (i.e., is treated as intra-category variation) may, in fact, contribute to differences in meaning in the foreign language. Thus the influence of the mature phonological system on the perception of foreign sounds is an artifact of how speech perception functions in general.
The role of categorization extends beyond recognition to production. So our perception of the language we hear will affect our ability to acquire and produce that language.
I have to say that L1 does affect L2 phoneme acquisition greatly. As mentioned in the article, this explains why different L1 learners would substitute different L1 sounds for a given L2 sound. The most common example as given by the author is the /θ/ in the English phoneme. For me, as an Indonesian native speaker, this particular sound and also the /r/ are the most challenging to produce. And it is interesting to see that it is not just me who has this problem, but all Indonesian speakers struggle to pronounce the same /θ/ and /r/ correctly in English. Most of Indonesians would substitute /t/ for /θ/ just like Russian speakers. When I first learned Japanese, I was very surprised to learn that to the Japanese, the English /θ/ sounds more like /s/. I just could not understand it because all my life I had always thought /θ/ to sound like /t/. In addition, Indonesian speakers have no difficulty in distinguishing /l/ and /r/ in English, while Japanese speakers do. I think these are just a very small and simple evidence that L1 does interfere L2 phonological acquisition.
I was intrigued by Brown’s results, especially the third experiment that proved her theory that L2 phonological acquisition in adults does not change over time /l-r/. She also pointed out that the L1 can either block the perception (e.g. /l-r/) or facilitate their perception if something similar exists in the L1. Brown wonders if there is any effect on increased linguistic input and this is why the third experiment contrasts low-level and high-level students living in Japan. I wonder if she would have had different results with a group of students living and studying in an english speaking country for the same amount of time? Since exposure may have an impact would the /l-r/ still be low?
The idea of categorization is apparent in all areas of linguistics which I think is very intriguing. It seems like our mind likes to generalize things to make room to store more knowledge.
In the article by Cynthia Brown, the recognition of the role of categorization in the process of distinguishing phonemes was the information that amazed me most. I found it fascinating that our mind can so easily “reorganize” sounds into these generalizations and then take a step further to filter out variations in the acoustics that do not differentiate meaning- all just based on our own language’s grammar!
For me, it is so interesting that Brown was able to come to the understanding that because of the filtering out of variations in phonemic categories not necessary for one’s native language, this categorical perception leads to L2 learner’s disability in differentiating the meanings of some words in foreign languages.
I agree with Alyssa’s statement above that while learning a new language, it would be helpful to take a phonetics course in order to be familiar with all of the different types of sounds used throughout all languages.
In this article, Brown discussed the differences in L2 acquisition, specifically, the acquisition of phonological systems in the L2. I found the experiment of L1 Japanese perception of the r- and l sounds to be very interesting. Reading about this experiment reminded me of the video we watched in class where we watched the mouth move while the same sound was being uttered. I agree with Alyssa, that taking a phonetics course when beginning a foreign language could be very useful. Many language programs do not have their majors take a phonetics course until most the language courses have already been complete, so I would be interested to see if those who took the class sooner performed better.
In reading Cynthia Brown’s article on phonological acquisition in infants and adults I was especially interested to learn about Feature Geometry theory. According to this theory it would seem that Adult phonological acquisition in an L2 is mediated more by L1 phonological categories than the presence or absence of UG. The idea that people lose access to their UG as they age never completely made sense to me because even though learning new languages and phonemes becomes increasingly difficult as we age, it is still not impossible, at least for some, suggesting that we do have at least some access to UG. It makes more sense to me that adults have difficulty pronouncing, and to a lesser degree differentiating between, new phonemes absent from their L1. The fact that adults have an L1 of phonemes to build from or compare to means that they will try to relate new phonemes to those they already possess and are attuned to hearing. I agree with Erica that these factors could be better addressed through explicit phonetics instruction at the beginning of learning an L2 in order to better attune learners to phonemes they are unfamiliar with.
Successful acquisition of phonological representations requires accurate perception of phonemic contrasts in the input. To Brown’s questions, I would add, what are the measures of phonological acquisition? In my opinion, it should be oral production, but also comprehension (which goes one step further than perception). The Speech Learning Model (SLM) maps L2 sounds on L1 sounds based on their proximity. In teaching language, teachers could use interlinguistic minimal pairs to highlight the subtle differences between the two speech sounds. In my experience with French for example, this could be achieved by opposing interlingual minimal pairs such as “mere†and “mirent†or “peek†and “piqueâ€. I also noticed that orthographemic interference is often a factor in phonological acquisition and thus is present for most adults but in addition, it also occurs in recently literate children.
I found this article very interesting. I had heard in other classes that acquisition of L2 phonemes that are similar to L1 phonemes was more difficult in terms of producing a native-sounding segment, but not much more detail was given and perception was not addressed. It sounded somewhat similar to the Speech Learning Model she mentioned. It is vague though as far as what is “similar” and “new”. The experiments that she did to how that it is the features of the phonemes, not the phonemes themselves, that make the difference in whether they can be acquired were fairly convincing. Though I would imagine that it would make a difference if the phonemes of one language are in an allophonic relationship in the other or simply not present. She mentioned that l/r are both a flap in Japanese so the fact that they are perceived categorically is not a surprise. The evidence from the other L1s she tested still makes me think she is right that features not present in the L1 will hinder acquisition of phonemes in the L2 that depend on that feature, and L1 features that are present will help acquisition of L2 phonemes that rely on those same features. It just seems it would make it that much more difficult if sounds that in one language are phonemic are allophonic in another, compared to sounds that are phonemic in one language but entirely absent in the other.
The most interesting aspect of this article for me was that it discusses that L2 phoneme acquisition is affected not only by language acquisition, but phonological representation, and speech perception as well.
I have never thought about much about speech perception affecting speech production before. Of course logically it makes sense because if you can’t hear a difference between two sounds, how are you supposed to produce that difference?
The article says that L2 learners try and categorize phonemes based upon the ones they already recognize in their L1. And there is an order that they categorize them in based upon certain features. It also says that as infants we have a greater ability to distinguish contrasting sounds in all languages but as we get older this ability begins to degrade from all potential contrasts to only native ones. This is explained by the fact that when we filter out only the contrasts that create different meanings in our native language, it makes processing much faster.
For me thinking about this raised the question of how easy is it for L2 learners to start listening for contrastive sounds again after they are exposed to a new language?
It also makes me think that in a way perhaps when learning a new language it is good to take a phonetics course in that language to try and at least become aware of the sounds that are produced differently in that language so that you can better try and perceive the differences.
One more thing that interested me is that I have a professor who is partially deaf, but she can speak at least 3 languages fluently. This makes me wonder if speech perception really matters all that much because someone who’s ability to perceive speech is impaired, but is still very successful at producing the different phonemes of different languages.
As she says in this article, I can say it is true that for Japanese it is hard to distinguish some English sounds such as l and r, or s and th. When we describe English words contain sounds of th and s such as “gas” and “path”, we use the same character for “s” and “th” part because we don’t have way to express the sound perfectly. So, we almost have no chance to know that there is a difference between the sound “s” and “th” and it leads our lack of ability to distinguish them. Also, most of Japanese learn English by using Japanese in school, so that they may naturally acquire English based upon their first language, Japanese. I wonder that if we learn L2 by using only the language, our phonological acquisition can be better or not.
Two interesting and correct elements that shape the phonological acquisition for non-native speakers; the factor that determines the phonological acquisition is the L1 features and not the segments, and the other fact that speakers’ ability to acquire novel contrasts which do not exist in their L1 grammar does not change over time or by developing their L2. I applied these elements on the Turkish language speakers who are learning or speaking English language. Turks are unable to produce the sounds /w/, /θ/, and /ð/. Same situation applies when producing /b/ instead of /p/ by the Arabic language speakers, due to the fact that these contrasts do not exist in their native L1. But accuracy in performance was different, as the Turkish speakers are unable to produce these contrasts regardless of the number of years of exposure to English language, but Arabic speakers were significantly developed their performance over time.
While reading this article, I was reminded of the exercise we did in class, trying to distinguish /p^h/, /b/, and /p/ sounds, as well as d, t, and th sounds (I can’t remember what language we listened to…). With that exercise, we talked about our cognitive “categorization skills” determine how we differentiate these sounds. In this article, this categorical perception is pointed out to perhaps be a hindrance to second language acquisition :
Yet although categorical perception aids in processing of one’s native language, it can be a barrier to correctly perceiving and processing a foreign language: variation in the acoustic signal which is filtered out by the native phonological system (i.e., is treated as intra-category variation) may, in fact, contribute to differences in meaning in the foreign language. Thus the influence of the mature phonological system on the perception of foreign sounds is an artifact of how speech perception functions in general.
The role of categorization extends beyond recognition to production. So our perception of the language we hear will affect our ability to acquire and produce that language.
I have to say that L1 does affect L2 phoneme acquisition greatly. As mentioned in the article, this explains why different L1 learners would substitute different L1 sounds for a given L2 sound. The most common example as given by the author is the /θ/ in the English phoneme. For me, as an Indonesian native speaker, this particular sound and also the /r/ are the most challenging to produce. And it is interesting to see that it is not just me who has this problem, but all Indonesian speakers struggle to pronounce the same /θ/ and /r/ correctly in English. Most of Indonesians would substitute /t/ for /θ/ just like Russian speakers. When I first learned Japanese, I was very surprised to learn that to the Japanese, the English /θ/ sounds more like /s/. I just could not understand it because all my life I had always thought /θ/ to sound like /t/. In addition, Indonesian speakers have no difficulty in distinguishing /l/ and /r/ in English, while Japanese speakers do. I think these are just a very small and simple evidence that L1 does interfere L2 phonological acquisition.
I was intrigued by Brown’s results, especially the third experiment that proved her theory that L2 phonological acquisition in adults does not change over time /l-r/. She also pointed out that the L1 can either block the perception (e.g. /l-r/) or facilitate their perception if something similar exists in the L1. Brown wonders if there is any effect on increased linguistic input and this is why the third experiment contrasts low-level and high-level students living in Japan. I wonder if she would have had different results with a group of students living and studying in an english speaking country for the same amount of time? Since exposure may have an impact would the /l-r/ still be low?
The idea of categorization is apparent in all areas of linguistics which I think is very intriguing. It seems like our mind likes to generalize things to make room to store more knowledge.
In the article by Cynthia Brown, the recognition of the role of categorization in the process of distinguishing phonemes was the information that amazed me most. I found it fascinating that our mind can so easily “reorganize” sounds into these generalizations and then take a step further to filter out variations in the acoustics that do not differentiate meaning- all just based on our own language’s grammar!
For me, it is so interesting that Brown was able to come to the understanding that because of the filtering out of variations in phonemic categories not necessary for one’s native language, this categorical perception leads to L2 learner’s disability in differentiating the meanings of some words in foreign languages.
I agree with Alyssa’s statement above that while learning a new language, it would be helpful to take a phonetics course in order to be familiar with all of the different types of sounds used throughout all languages.
In this article, Brown discussed the differences in L2 acquisition, specifically, the acquisition of phonological systems in the L2. I found the experiment of L1 Japanese perception of the r- and l sounds to be very interesting. Reading about this experiment reminded me of the video we watched in class where we watched the mouth move while the same sound was being uttered. I agree with Alyssa, that taking a phonetics course when beginning a foreign language could be very useful. Many language programs do not have their majors take a phonetics course until most the language courses have already been complete, so I would be interested to see if those who took the class sooner performed better.
In reading Cynthia Brown’s article on phonological acquisition in infants and adults I was especially interested to learn about Feature Geometry theory. According to this theory it would seem that Adult phonological acquisition in an L2 is mediated more by L1 phonological categories than the presence or absence of UG. The idea that people lose access to their UG as they age never completely made sense to me because even though learning new languages and phonemes becomes increasingly difficult as we age, it is still not impossible, at least for some, suggesting that we do have at least some access to UG. It makes more sense to me that adults have difficulty pronouncing, and to a lesser degree differentiating between, new phonemes absent from their L1. The fact that adults have an L1 of phonemes to build from or compare to means that they will try to relate new phonemes to those they already possess and are attuned to hearing. I agree with Erica that these factors could be better addressed through explicit phonetics instruction at the beginning of learning an L2 in order to better attune learners to phonemes they are unfamiliar with.
Successful acquisition of phonological representations requires accurate perception of phonemic contrasts in the input. To Brown’s questions, I would add, what are the measures of phonological acquisition? In my opinion, it should be oral production, but also comprehension (which goes one step further than perception). The Speech Learning Model (SLM) maps L2 sounds on L1 sounds based on their proximity. In teaching language, teachers could use interlinguistic minimal pairs to highlight the subtle differences between the two speech sounds. In my experience with French for example, this could be achieved by opposing interlingual minimal pairs such as “mere†and “mirent†or “peek†and “piqueâ€. I also noticed that orthographemic interference is often a factor in phonological acquisition and thus is present for most adults but in addition, it also occurs in recently literate children.