You are here: Home Q&A Questions and Answers

Questions and Answers

Vragen en Antwoorden Nederlands Fragen und Antworten Deutsch

Is there something you have always wanted to know about language? We might have an answer! On this page we answer questions about various aspects of language asked by people outside of the language researcher community.

Show or Hide answerWhat is the similarity between a natural languages and programming languages?
Link & Share

Some programming languages certainly look a lot like natural languages.  For example, here’s some Python code that searches through a list of names and prints one out if it’s also in a list called invited_people.

Python

for name in my_list:
  if name in invited_people:
    print name

However, some other programming languages are much less readable.  Here is some code in Scheme that does the same as the code above.


Scheme

(map (lambda (name) (if (cond ((member name invited_people) name)) (display name) name)) my_list)

So, how similar or different are natural languages and programming languages really? To answer these questions, we need to understand some central terms that linguists use to describe the structure of languages, rather than just looking at the surface of what looks similar or not. 

Code

 

Similar structures of semantics and syntax

Two of the most central concepts in linguistics are the concepts of semantics and syntax. In short, semantics is the linguistic term for meaning, but a more precise explanation is that semantics contains the information connected to concepts. For instance, a word form like “sleep” (spelled S-L-E-E-P, in either letters or pronounced as sounds) designates a certain action of a living organism and that’s the semantics of that word. The syntax on the other hand is the structure of how words of different kinds (e.g. nouns and verbs) can be combined and inflected. The sentence “My ideas sleep” is a well-formed English sentence from the point of view of syntax, but the semantics of this sentence is not well-formed since ideas are not alive and thus cannot sleep. So, semantics and syntax have rules, but semantics relates to meaning and syntax relates to how words can be combined.

Having explained the semantics and syntax of natural languages, let’s turn back to programming. In programming languages, the coder has an intention of what the code should do. That could be called the semantics or the meaning of the code. The syntax of the programming language links a certain snippet of code, including its “words” (that is variables, functions, indices, different kinds of parenthesis etc.) to the intended meaning. The examples of python and scheme above have the same semantics, while the syntax of the two programming languages differ. 

Different purposes

We have described many parallels between the basic structure of natural languages and programming languages, but how far does the analogy go? After all, natural languages are shaped by communicative needs and the functional constraints of human brains. Programming languages on the other hand are designed to have the capacities of a Turing machine, i.e. to do every computation that humans can do with pen and paper, again and again.

 

It is necessary for programming languages to be fixed and closed, while natural languages are open-ended and allow blends. Code allows long lists of input data to be read in, stored and rapidly parsed by shuffling around data in many steps, to finally arrive at some output data. The point is that this is done in a rigorous way. Natural languages on the other hand must allow their speakers to greet each other, make promises, give vague answers and tell lies. New meanings and syntax constantly appear in natural languages and there is a gradual change of e.g. word meanings. A sentence from a spoken language can have several possible meanings.  For example, the sentence “I saw the dog with the telescope”, has two possible meanings (seeing a dog through a telescope or seeing a dog that owns a telescope). People use context and their knowledge of the world to tell the difference between these meanings. Natural languages thus depend on an ever changing culture, creating nuances and blends of meanings, for different people in different cultures and contexts. Programming languages don’t exhibit this kind of flexibility in interpretation. In programming languages, a line of code has a single meaning, so that the output can be reproduced with high fidelity.

Answer by: Julia Udden, Harald Hammarström and Rick Jansen

Show or Hide answerHomophones – what are they and why do they exist at all?
Link & Share

Homophones are words that sound the same but have two or more distinct meanings. This phenomenon occurs in all spoken languages. Take for instance the English words FLOWER and FLOUR. These words sound the same, even though they differ in several letters when written down (therefore called heterographic homophones). Other homophones sound the same and also look the same, such as the words BANK (bench/river bed) and BANK (financial institution) in both Dutch and English. Such words are sometimes called homographic homophones. Words with very similar sounds but different meanings also exist between languages. An example is the word WIE, meaning 'who' in Dutch, but 'how' in German.

1.07

One might think that homophones would create serious problems for the hearer or listener. How can one possibly know what a speaker means when she says a sentence like "I hate the mouse"? Indeed, many studies have shown that listeners are a little slower to understand ambiguous words than unambiguous ones. However, in most cases, it is evident from the context what the intended meaning is. The above sentence might for example appear in the contexts of "I don't mind most of my daughter's pets, but I hate the mouse" or "I love my new computer, but I hate the mouse". People normally figure out the intended meaning so quickly, that they don't even perceive the alternative. Preceding linguistic context and common world knowledge thus help us in understanding a speaker’s intended message.

Why do homophones exist? It seems much less confusing to have separate sounds for separate concepts. Linguists take sound change as an important factor that can lead to the existence of homophones. For instance, in the early 18th century the first letter of the English word KNIGHT was no longer pronounced, making it a homophone with the word NIGHT. Also language contact creates homophones. The English word DATE was relatively recently adopted into Dutch, becoming a homophone with the already existing word DEED. Some changes over time thus create new homophones, whereas other changes undo the homophonic status of a word. Now the Dutch verb form ZOUDT (would) is no longer commonly used, the similarly sounding noun ZOUT (salt) is losing its homophonic status.

Finally, a particularly nice characteristic of homophones is that they are often used in puns or as stylistic elements in literary texts. In Shakespeare’s Romeo and Juliet (Act I, Scene IV, line 13-16), Romeo for instances uses a homophone when he refrains from following up his friend Mercutio‘s advice to dance:

Mercutio:              Nay, gentle Romeo, we must have you dance.

Romeo:                Not I, believe me: you have dancing shoes

                          With nimble soles: I have a soul of lead

                          So stakes me to the ground I cannot move.

Among other things, it is such elegant use of homophones that has led to Shakespeare’s literary success.

 By David Peeters and Antje S. Meyer

Further reading:

Bloomfield, L. (1933). Language. New York: Henry Holt and Company.

Cutler, A., & Van Donselaar, W. (2001). Voornaam is not (really) a homophone: Lexical prosody and lexical access in Dutch. Language and speech, 44(2), 171-195. (link)

Rodd, J., Gaskell, G., & Marslen-Wilson, W. (2002). Making sense of semantic ambiguity: Semantic competition in lexical access. Journal of Memory and Language, 46(2), 245-266. (link)

Tabossi, P. (1988). Accessing lexical ambiguity in different types of sentential contexts. Journal of Memory and Language, 27(3), 324-340. (link)

Show or Hide answerWhat do researchers mean when they talk about "maternal language"?
Link & Share

When researchers talk about "maternal language", they talk about child-directed language. Modern research on the way in which caregivers talk to children started in the late seventies. Scholars who study language acquisition were interested in understanding how language learning is influenced by the way caregivers talk to their children. Since the main caregivers were usually mothers, most first studies focused on maternal language (often called Motherese), which was usually described as having higher tones, a wider tonal range and a simpler vocabulary. However, we now understand that the ways in which mothers modify their speech for their children are extremely variable. For example, some mothers use a wider tone range than usual, but many mothers use nearly the same tone range they use with adults. Mothers’ speech also changes to adapt to their children’s language abilities, so maternal speech can be very different from one month to the next.

Q&A maternal 6.2

Image: Eoin Dubsky

Also, in some cultures mothers use little or no “baby talk” when interacting with their children. Because of this, it is impossible to define a universal maternal language. It’s also important to remember that fathers, grandparents, nannies, older siblings, cousins, and even non-family members often modify their speech when they address a young child. For this reason, researchers today prefer to use the label “child-directed speech” instead of "maternal language" or “motherese”. Some recent studies also show that we do more than just modify our speech for children: we also modify our hand gestures and demonstrative actions, usually making them slower, bigger, and closer to the child. In sum, child-directed language seems to be only one aspect of a broader communicative phenomenon between caregivers and children.

Emanuela Campisi, Marisa Casillas & Elma Hilbrink

Further Reading:

Fernald, A., Taeschner, T., Dunn, J., Papousek, M., de Boysson-Bardies, B., & Fukui, I. (1989). A cross-language study of prosodic modifications in mothers' and fathers' speech to preverbal infants. Journal of Child Language, 16, 477–501.

Rowe, M. L. (2008). Child-directed speech: relation to socioeconomic status, knowledge of child development and child vocabulary skill. Journal of Child Language35, 185–205.

Show or Hide answerWhat is the difference between surface and deep layer in language?
Link & Share

The terms "surface layer" and "deep layer" refer to different levels that information goes through in the language production system. For example, imagine that you see a dog chasing a mailman. When you encode this information, you create a representation that includes three pieces of information: a dog, a mailman, and the action chasing. This information exists in the mind of the speaker as a "deep" structure. If you want to express this information linguistically, you can, for example, produce a sentence like "The dog is chasing the mailman." This is the "surface" layer: it consists of the words and sounds produced by a speaker (or writer) and perceived by a listener (or reader). You can also produce a sentence like "The mailman is being chased by a dog" to describe the same event -- here, the order in which you mention the two characters (the "surface" layer) is different from the first sentence, but both sentences are derived from the same "deep" representation. Linguists propose that you can perform movement operations to transform the information encoded in the "deep" layer into the "surface" layer, and refer to these movement operations as linguistic rules. Linguistic rules are part of the grammar of a language and must be learned by speakers in order to produce grammatically correct sentences.

Chomsky

Image: Duncan Rawlinson

Rules exist for different types of utterances. Other examples of rules, or movement operations between "deep" and "surface" layers, include declarative sentences (You have a dog) and their corresponding interrogative sentences (Do you have a dog?). Here, the movement operations include switching the order of the first two words of the sentence.

by Gwilym Lockwood & Agnieszka Konopka

Further reading:

Chomsky, N. (1957). Syntactic Structures. Mouton.
Chomsky, N. (1965). Aspects of the Theory of Syntax. MIT Press.

About MPI

This is the MPI

The Max Planck Institute for Psycholinguistics is an institute of the German Max Planck Society. Our mission is to undertake basic research into the psychological,social and biological foundations of language. The goal is to understand how our minds and brains process language, how language interacts with other aspects of mind, and how we can learn languages of quite different types.

The institute is situated on the campus of the Radboud University. We participate in the Donders Institute for Brain, Cognition and Behaviour, and have particularly close ties to that institute's Centre for Cognitive Neuroimaging. We also participate in the Centre for Language Studies. A joint graduate school, the IMPRS in Language Sciences, links the Donders Institute, the CLS and the MPI.

 
Questions and Answers

whiet question mark on MPG green 124pt, stroke 2pt

This project was coordinated by:

Katrien Segaert 
Katerina Kucera
Judith Holler

Sean Roberts
Agnieszka Konopka
Gwilym Lockwood
Elma Hilbrink
Joost Rommers
Mark Dingemanse
Connie de Vos