They’re Not Random: Understanding the Patterns Behind Multilingual Learners’ Errors

When multilingual learners produce language differently from monolingual speakers, those differences are often viewed only through the lens of what students are missing.

A student writes:

She likes to play wif me.

Another student says:

Yesterday I goed to the store.

A teacher notices a student repeatedly saying:

My brother play soccer.

At first glance, these may look like unrelated mistakes. However, decades of research in second language acquisition suggest something far more interesting: many of these errors are signs that students are actively constructing a new language system. These errors do not simply reveal what students are missing. They often reveal what students already know. To understand why, educators need to understand four important principles about multilingual language development.

Principle 1: Students Use What They Already Know About Language

Students do not approach a new language as blank slates. They bring with them an already developing language system that influences how they perceive, process, and produce additional languages. Researchers refer to this interaction between languages as cross-linguistic influence.

Early theories of second language acquisition, such as Robert Lado’s Contrastive Analysis Hypothesis (1957), suggested that comparing two languages could predict where learners would experience ease or difficulty. Although researchers eventually discovered that language development is far more complex, contrastive analysis remains a useful tool because similarities and differences across languages often explain predictable patterns.

Consider phonology—the sound system of language. Common sense tells us that sounds present in one language but absent in another may be harder for learners to perceive and produce. But the reason why this happens is fascinating.

During infancy, the brain becomes increasingly specialized for the sounds it hears regularly. Early in development, infants can distinguish sound contrasts across many different languages. Over time, however, their phonological systems become tuned to the language or languages in their environment.

An illuminating example comes from research on Japanese infants and the English /l/ and /r/ distinction. At around 5–6 months of age, Japanese infants can distinguish between these two sounds, even though this distinction does not exist in Japanese. However, by around one year of age, infants exposed primarily to Japanese become significantly less sensitive to this contrast because their phonological system has specialized according to the language input around them (Tsushima et al., 1994).

This specialization is incredibly efficient. It allows children to process their home language rapidly and automatically. However, when learning an additional language, students may need to develop new phonological categories or refine existing ones.

For example, Spanish has a smaller vowel inventory than English. As a result, Spanish-speaking students learning English may initially have difficulty perceiving and producing certain English vowel sounds.

A student may hear or produce:

bit → beet

or

but → boot

These productions are not random. They reflect the student using an existing phonological system while learning the sound patterns of a new language.

Of course, understanding why these patterns occur does not mean educators should ignore them. On the contrary, multilingual learners benefit from explicit instruction that helps them notice, process, and produce sounds that may not exist in their home language. For example, the /th/ phoneme, as in the beginning sound of the word thread, does not exist in most dialects of Spanish. Therefore, Spanish-speaking students learning English often benefit from instruction that explicitly draws their attention to how this new sound is produced.

Teachers can model the sound while helping students notice important articulatory features, such as place of articulation (where the sound is produced), manner of articulation (how airflow moves through the mouth), and voicing (whether the vocal folds vibrate). For example, students can observe that producing /th/ requires placing the tongue between the teeth while allowing air to flow continuously. Using a small mirror can help students see the position of their tongue, teeth, and lips, while placing a hand near the mouth can help them feel the airflow.

Similarly, placing a hand on the throat can help students notice whether sounds are voiced or unvoiced.

However, effective instruction should not focus only on differences between languages. Teachers should also highlight important similarities that allow students to leverage what they already know. For example, Spanish and English both contain the /k/ phoneme. A Spanish-speaking student who can hear and produce the first sound in casa does not need to learn an entirely new sound when reading an English word like cat, although very slight differences in pronunciation do exist. The student already has the phonological category. What they need to learn is how that familiar sound maps onto the spelling patterns of the new language.

Educators interested in learning more about cross-linguistic connections within phonology and phonics instruction across English and Spanish might benefit from reading this article.

Principle 2: Errors Reveal What Students Understand

While cross-linguistic influence explains some patterns, it does not explain everything. Multilingual learners do not simply copy patterns from one language into another. They actively construct a developing language system.

Second language researcher Pit Corder (1976) made an important distinction between mistakes and errors. Mistakes are temporary performance issues—something a learner knows but does incorrectly in the moment. Errors, however, can reveal something about the learner’s underlying language system.

Consider a child who says:

Yesterday I goed to the store.

It would be easy to assume that this means the student does not know past tense. But the opposite may actually be true. This sentence reveals that the student has discovered an important grammatical pattern: English often marks past events by adding -ed to verbs. The student is not failing to use a rule. The student is applying a rule too broadly.

Building on this idea, Larry Selinker (1972) introduced the concept of interlanguage—the developing linguistic system created by someone learning an additional language. This system is influenced by the learner’s existing language and the new language, but it is also its own unique system. In other words, many errors are not signs of confusion. They are signs of development.

Understanding this does not mean educators ignore errors. Rather, errors provide valuable information about how students are currently organizing the language they are learning. By analyzing these patterns, teachers can provide instruction and feedback that align with students’ current stage of language development and move them toward more advanced language use.

Principle 3: Language Structures Develop Over Time

Another reason multilingual learners make predictable errors is because not all parts of language are acquired at the same time. Research on grammatical development has consistently found that learners often acquire certain structures before others. We can see evidence of this in studies examining the acquisition of English morphemes by second language learners (see Dulay & Burt 1973; Bailey et al., 1974).

A morpheme is the smallest unit of meaning in a language. For example, the word dogs contains two morphemes: the base morpheme dog, which carries the main meaning, and the plural -s, which communicates that there is more than one dog. Similarly, the word walks contains the base morpheme walk and the third-person singular -s, which marks agreement with the subject.

Interestingly, studies of second-language development have found that English learners, regardless of their home language, acquire plural -s before third-person singular -s.

A student may correctly say:

three dogs

while still saying:

The man walk every day.

Although both examples involve adding an -s morpheme, the underlying grammatical demands are not the same.

The plural -s requires the learner to mark information within a noun phrase:

However, third-person singular -s requires coordinating information across different parts of the sentence (i.e., the noun phrase and the verb phrase):

The learner must identify the subject, process that the subject is third-person singular, and then mark that information on the verb—a task requiring a higher stage of processing ability and substantial cognitive resources during real-time communication (Pienemann, 1998). Additionally, the third-person singular -s carries information that is somewhat redundant. The subject the man already communicates that we are talking about one person who is neither the speaker nor the listener. The -s on walks repeats this grammatical information by marking the verb for third-person singular agreement (VanPatten, 2004).

Because this relationship requires learners to coordinate information across different parts of the sentence—and because the meaning of the sentence can often still be understood without the marker—it is not surprising that third-person singular -s tends to be acquired later.

Therefore, when a multilingual learner says The man walk every day—even when the learner demonstrates relatively advanced English proficiency in other areas—the missing -s does not automatically indicate a language disorder or even a lack of understanding. Instead, it may reflect a predictable pattern in language development.

An important nuance is worth mentioning here. Although the third-person singular -s is often acquired later by learners of English, this does not mean educators should ignore it or other common errors during instruction. During reading and writing activities, teachers should still draw attention to this grammatical feature and provide corrective feedback so students develop more explicit awareness of how and when it is used. This explicit knowledge can support students, especially during writing, where they have more time to monitor and revise their language.

However, educators should understand that consistently producing the third-person singular -s during spontaneous communication may take longer because learners must first have internalized the appropriate processing procedure, and then, they must apply it while simultaneously managing all the other demands involved in communicating.

Principle 4: Bilingual Students Use Their Full Linguistic Repertoire

A Spanish-English bilingual student might say, “Voy al supermarket,” not necessarily because they forgot the Spanish word supermercado, but because bilingual speakers often strategically draw from their full linguistic repertoire. This practice, often called code-switching or translanguaging, occurs when bilingual individuals use elements from multiple languages during communication.

Importantly, code-switching is not random. Research has repeatedly demonstrated that bilingual speakers who have acquired proficiency in multiple languages move between those languages in systematic and rule-governed ways. In fact, bilingual speakers often follow complex grammatical constraints without any conscious awareness they are doing so.

Consider the two sentences below:

1. *Yo eat hamburgers every week.

   (I eat hamburgers every week.)

   
   

2. Mis amigos y yo eat hamburgers every week.

   (My friends and I eat hamburgers every week.)

Interestingly, many Spanish-English bilingual speakers judge the first sentence as unnatural while accepting the second sentence as something a bilingual speaker might produce. Why?

The answer is that code-switching is constrained by the underlying grammatical systems of the languages involved. Although both examples contain the Spanish pronoun yo next to the English verb eat, the grammatical relationship between these words is different. In the first sentence, yo functions by itself as the subject directly connected to the English verb phrase. This creates a conflict at a very small level of grammatical structure where researchers have found code-switching is often restricted. In the second sentence, however, yo is not functioning independently. It is part of the larger phrase mis amigos y yo (“my friends and I”). Because the switch occurs between larger grammatical units rather than inside a tightly connected structure, bilingual speakers generally find this type of sentence acceptable (see van Gelderen & MacSwan 2008 for a technical discussion of this code-switching phenomenon).

Examples like these demonstrate an important point: bilingual students are not randomly combining pieces of different languages. Their language use reflects sophisticated knowledge of multiple linguistic systems. In other words, code-switching does not mean bilingual students are confused or unable to separate languages. Instead, it often reflects their ability to strategically use their complete linguistic repertoire.

Seeing Errors Differently

Understanding multilingual development changes the way educators respond. The important question is not: “Does this student make language errors?” All developing language learners make errors. Instead, we should ask:

• Is this pattern expected given the languages the student knows?

• Is this pattern typical given what we know about language development?

• Is this typical given their amount of exposure?

• Does the student respond when provided instruction and feedback?

• Are similar difficulties present across languages?

  
  

These questions help educators avoid two common mistakes. We do not want to identify normal multilingual development as a disability. But we also do not want to overlook genuine language and reading difficulties simply because a child is multilingual. The goal is not to ignore errors. The goal is to understand what those errors are telling us and respond accordingly.

In the next post, we will explore these questions in greater depth and examine how educators can better distinguish expected patterns of multilingual development from learning difficulties that may require additional support.

References

Bailey, N., Madden, C., & Krashen, S. (1974). Is there a “natural sequence” in

adult second language learning? Language Learning, 24(2), 235-243.

Corder, S.P. (1967). The significance of learner's Errors. IRAL, 5(1-4), 161-170.

Dulay, H. C., & Burt, M. K. (1973). Should we teach children syntax? Language Learning, 23(2), 245-258.

Lado, R. (1957). Linguistics across cultures: Applied linguistics for teachers. Ann Arbor: University of Michigan Press.

Pienemann, M. (1998). Language, processing and second language development: Processability theory. Amsterdam, Netherlands: John Benjamins.

Selinker, L. (1972). Interlanguage. International Review of Applied Linguistics in Language Teaching, 10(3), 209-231. 

T. Tsushima, O. Takizawa, M. Sasaki, S. Shiraki, K. Nishi, M. Kohno, P. Menyuk and C. Best, ‘‘Discrimination of English /r– l/ and /w–y/ by Japanese infants at 6–12 months: Language specific developmental changes in speech perception abilities,’’ 1994 International Conference on Spoken Language, Vol. S28F-1, pp. 1695–1698 (Yokohama, Japan, 1994).

Van Gelderen, E., & MacSwan. J. (2008). Interface conditions and codeswitching: An F-movement analysis of pronouns and lexical DPs. Lingua 118(6), 765-776.

VanPatten, B. (2004). Input processing in second language acquisition. In B. VanPatten (Ed.), Processing instruction: Theory, research, and commentary (pp. 5-31) Mahwah, NJ: Lawrence Erlbaum.