The purpose of this study is to evaluate the effectiveness of learning a new language with new characters, in particular Chinese characters, in other words, how do second language adult students acquire knowledge about orthographic characters of a new language system. While past research has shown that adult learners with no previous Chinese exposure learn more effectively by receiving explicit instructions of orthographic rules (Taft & Chung, 1999; Dunlap, Perfetti, Liu & Wu). This study focuses particularly in the semantic rules. Our results support previous results in that those learners who were given explicit instructions of semantic rules did learn more effectively the new orthographic structures compared to those who only received implicit instructions.
This study only focuses on the Semantic radicals, and how effectively they are learned by naive foreign learners. Chinese children learn better when they know the semantic radical rules, but these are not normally taught to foreign students explicitly. It might be because it can create confusion. There are many characters that are pronounced in a similar way and this can be a cause of confusion as more characters are learned.
One study about different teaching approaches of Chinese explains that there are two types of approaches when teaching characters: centered approach, or the meaning centered approach. In the meaning centered approach learners read and naturally learn to recognize the characters while, they are reading. The problem is that once they start to learn more characters, due to the similarity in pronunciations, they start to get confused. Thus, a character centered approach is much more suitable and effective (Lam, 2011). They learn the structure of each character and the rules that apply. In other words, they teach the semantic and phonetic rules of radicals.
This study goes in line with other findings of character centered approaches. As characters are compound characters made of radicals, a character centered approach would involve distinguishing the radicals and knowing the rules. While adult learners can indeed pick up semantic radical knowledge through implicit learning when being exposed, however, explicit instructions on semantic radicals have proven to me even more effective.
For example DeKeyser, (2008) analyzed many cognitive psychology research done on implicit and explicit learning of new language systems, in particular Chinese. His conclusion was that cognitive psychologists can still not provide consistent and sound evidence that learners of a new language system can learn characters without being aware of them. In other words, explicit learning is still more effective, and in practice it goes as well hand by hand with the character centered teaching approach mentioned above.
There are many characters to be learnt, all new and abstract. The phonetics do not help, as many have very similar phonetics. While memory and implicit learning through exposure and use of the characters can indeed be done, as shown by previous research, to be able to retain over time, the semantic knowledge is important indeed.
Taft and Chung, (2009) did an experiment to evaluate how helpful would be for learners to know about the internal radical structure of the characters in order to help them memorize them. The results are consistent with previous results: explicit instruction about the radical rules is more effective than not doing it. By having an awareness of the radicals even if the exposure has been short, the memory of them lasts for at least a week.
A more recent study to investigate the effects of radical awareness on foreign students learning Chinese characters, created a learning platform. The results showed that orthographic awareness when teaching is the most effective approach (Chen & et al., 2003).
Perfetti and Wang (2004) showed in their study that learners of Chinese who do not receive explicit instruction about visual-orthographic structures of the characters do acquire implicitly the rules by means of constant exposure to them. He compared this learning process to Chinese children who are forced to speak and write and read, in other words, to do frequent use of them and through this process they learn effectively. But foreign learners do not have this benefit. They did a second experiment. These results showed that learning with explicit instructions gives more effective results, as it speeds up the learning process.
Finally another research has performed a manipulation of semantic subcomponents to see if these are reliable cues to word meanings, and whether the information should be taught implicitly or explicitly to show the most effective results. Their conclusion is that learners do benefit from being explicitly taught the connection between the semantic subcomponents of characters, and thus, the meaning of the words. This is because they are then able to apply this knowledge in new vocabulary and understand it as a whole. This second part of the process is the one that speeds up the learning process (Dunlap, Perfetti, Liu, & Wu., 2011).
The current study looks at early adult acquisition of Chinese – a writing system dramatically different from English. Its’ purpose is to find out if adult second language learners of Chinese can indeed learn the function role of semantic radicals. And if they do, which is the most effective way of learning them, implicitly or explicitly? The Hypothesis of this experiment is that those learners who have been explicit been given instructions about the radical rules and their semantic meaning have an advantage that will have a stronger effect in the long run. Those learners who received explicit instructions about orthographic rules will perform better not only in the first exposure test, but also weeks after the exposure.
The results show that explicit teaching of regular radicals is significantly more effective than allowing the learner to get to know them implicitly. However, the irregulars are better learned through implicit learning. These results are based on the Post Test results.
In the generalization test, which is the one that shows the effects of explicit learning to learn future new characters, as it shows new characters that were not previously shown, but with same radicals, shows however, that while the explicit learning group has learned better both same and different structures, the implicit learning group has done slightly better than the explicit group when learning different structures.
The Post test results of the Within Subjects Effects for regularity is also statistically significant while the regularity*Learning is non-significant. The Between Subjects for Learning has also been non-significant.
The Generalization Test results of Within-Subjects effects show that ‘Structure*Learning’ condition is statistically significant. The results of Between-Subjects effects of the ‘Learning’ condition for the same test has also shown to be statistically significant.
The results show it is statistically valid to confirm that explicit learning is more effective than implicit learning, even though in the irregularity condition implicit learning proves to increase, though not as much as explicit learning, compared to the learning of regular words.
1 Post Test - Mean Accuracy Results (%)
2 Generalization Test - Mean Accuracy Results (%)
Will teaching through explicit instructions about semantic rules, specifically semantic radicals, improve the learning of Chinese characters in the long run? Our hypothesis for this research is that indeed, those who have been given explicit instructions will come back after two weeks and take a test and will indeed do better. Memory can fade after a few weeks, but the understanding of the rule and applying that the rule might remain in time.
Furthermore, the understanding of the rule goes further than simply understanding the meaning of the character, but it helps to understand the meaning of the structure or sentence. Learners as we have mentioned already are able to understand the structure and sentence as a whole (Dunlap, Perfetti, Liu, & Wu., 2011).
Our results show that those learners who received explicit instructions did better in the post test for the regular characters than for the irregular. However those who were told to learn and memorize as best as they could do better in the irregular characters. One reason for this could be that not all rules might have applied in the irregular words. However learners who memorized them had more chances to know them as they did not know the different rules.
However, what is most important of all is the generalization test. This one is the one that would prove our hypothesis that explicit learning achieves better results on the long run. After two weeks of having done the first test, after the post test, learners were provided with a generalization test. New items were given to them, and this was a chance to prove how the rules could be applied to unknown vocabulary and overall meaning. They did not need to know the exact meaning as they were new items they have never seen before; they only had to know the radical semantic meaning. In a real base scenario, when the meaning of a word is not clear, the reader looks at the whole sentence, and tries to reach the meaning of an unknown word in this way.
Knowing the radical rules, and from there gather the overall or estimate meaning is a good comparison, analogically speaking with the example above. And those who indeed had received explicit instructions did better. Those who did not receive instructions could not possible know the semantic meaning as they did not have the rules. Their only aid was their memory about what the radical meant, but the connection of the radical with the whole word was left for them to learn implicitly.
The generalization results also support Perfetti and Wang, (2004) results in that while it is possible to learn implicitly through an intense exposure to the language, explicit explanation does speed up the process. The post test would be the one that say that one group learned more than another. But the generalization test would be the one that indicates that there is further learning to be done, the one that would indicate that the continuing learning process could be more efficient and thus faster, as Perfetti and Wang, 2004 found in their research. And indeed those who knew the rules did better indeed, indicating that, with further teaching, the speed would be much faster than those learning implicitly.
Most of the previous research has focused on immediate learning of Chinese characters and whether implicit or explicit learning was more effective. But they all analyzed the immediate tests and did a follow up test to still evaluate after some time, days or weeks, which group did better.
This study differentiates to the previous studies in that the analysis was done on the post test and the generalization test. Especially the last one provides great insight into how further knowledge would be taken by the learner. The way the semantic rules were applied makes a significant difference in the speed of future learning.
One minor detail but one that could potentially become a confounding variable, is be the fact that some participants knew other languages, more than one. It would be interesting to evaluate in which way knowing more languages could affect the way the new language is learned. Are there any advantages in knowing more than one language? How many participants that knew more than one language was part of the explicit group? Would Chinese be considered as a new language or not because of its complete new character system? If it did, it would be also interesting to evaluate whether the number of those participants had any effect on the results.
Cognitive processes are complex, in learning new Chinese characters there are many different processes going on. The current test showed that explicit teaching of semantic radicals gives good results over time, and especially can lead to a more effective and faster learning of future characters. When learning Chinese characters there are different components to be learnt at the same time: the phonetic, the abstract graph of the character, the semantic including the regularity or irregularity. Some of these elements necessarily require memory learning, such as the abstract form of the character, and this can only be done by memory using different strategies. These implicit memory strategies can be combined to aid the memory, with the explicit learning of the phonetic and semantic rules. The combination of implicit and explicit learning for the different parts of the Characters seems to be the ideal teaching approach. A study that could investigate this teaching strategy would be extremely interesting and useful.
Chen, H., Hsu, C., Chang, L., Lin, Y., Chang, K., & Sung, Y. (2013) Using a radical-derived character e-learning platform to increase learner knowledge of Chinese characters. Language Learning & Technology, 17, 89-106. Retrieved from http://llt.msu.edu/issues/february2013/chenetal.pdf
DeKeyser, R. (2008). Implicit and explicit learning. In C. J. Doughty & M.H. Long (Eds). The Handbook of Second Language Acquisition. pp. 313-348. Blackwell Publishing Ltd: Oxford, UK
Dunlap, S., Perfetti, C. A., Liu, Y., & Wu, S.-M. (2011). Learning vocabulary in Chinese as a foreign language: Effects of explicit instruction and semantic cue reliability. Retrieved from http://www.pitt.edu/~perfetti/PDF/DunlapLearningVocabulary.pdf
Lam, H. (2011). A critical analysis of the various ways of teaching Chinese characters. Electronic Journal of Foreign Language Teaching, 8, 57-70. Retrieved from http://e-flt.nus.edu.sg/v8n12011/lam.pdf
Taft, M., & Chung, K. (1999). Using radicals in teaching Chinese characters to second language learners. Psychologia, 42, 243-251. Retrieved from http://www2.psy.unsw.edu.au/Users/mtaft/TaftANDChung1999.PDF
Wang, M., Liu, Y., & Perfetti, C. (2004) The implicit and explicit learning of orthographic structure and function of a new writing system. Scientific Studies of Reading, 8, 357-379. Retrieved from http://www.pitt.edu/~perfetti/PDF/Implicit%20and%20explicit%20learning%20of%20ortho-%20Wang,%20Liu.pdf