The Romance Languages: A Linguistic Treasure Box


This week's guest blog post comes from Dr Sam Wolfe. Sam is a Romance dialectologist and historical linguist based at St John's College, University of Cambridge. He's interested in how and why languages change and asking how much varies and what is universal across human languages. He has particular interests in dialects spoken in Italy, the history of French and parallels between the Romance and Germanic languages.

Language learners come in all shapes and sizes, but we often have in common a certain sense of adventure and a desire to study and learn about things which are novel or unknown. At times this means we're tempted to ignore familiar languages on our doorstep in favour of those we view as more unusual, or even “exotic”. In this blog post I'd like to show that the Romance family of languages, descended from Latin, still have plenty of tricks up their sleeves to keep language learners on their toes.

Statements like “Once you know Spanish, Portuguese is easy” or “If you've studied Latin, you really know Italian and French” will be familiar to lots of us from school classrooms. There's a grain of truth in these statements, particularly when it comes to the core vocabulary of the languages in question. To take two simple examples, consider the words for “life” and “wolf” across a number of Romance languages. All have a common Latin root and the outcomes we find today are strikingly similar, even if certain sound changes have meant they are not identical:

Latin Italian Sardinian Spanish Catalan Portuguese French Gascon Romansh Romanian
vita vita vida vida vida vida vie vita vita viaţă
lÅ­pum lupo lupu lobo llop lobo loup lop luf lup

But when we start looking at the core grammatical structure of the language, we see that we can only push these ideas so far. Let's take one case study to illustrate this point: how we express events that have taken place in the past.

Anybody who has studied basic French at school, will remember (however distantly) that to express an event that took place at a particular point in the past you often use a construction known as the compound past, where a past participle (equivalent to English done, spoken, played) is combined with an auxiliary verb. This auxiliary verb is typically the verb for “have”, avoir. To express “I spoke” or “I have spoken” we therefore say j'ai parlé. For a certain class of verbs, however, we use the verb “to be” être as an auxiliary. “I have arrived” is therefore not j'ai arrivé but je suis arrivé, which literally translates as “I am arrived”.

At first glance, Standard Italian seems to have a similar system. We find ho parlato “I spoke, I have spoken” formed with the verb “have” and sono arrivato “I arrived, I have arrived” formed with “be”. But even here we don't find exactly the same system. Certain verbs in Italian such as fugire “to flee”, esistere “to exist” and sopravvivere “to survive” use essere “be” to form their past tense, whilst their French counterparts use avoir “have”. The take home message is that we can use our intuitions from the grammar of closely related languages to guide us in learning new ones, but we have to be a little cautious in applying them wholesale to a language that seems at first glance to be similar.

Elsewhere in the Romance languages, we find a wealth of different patterns which show us that the system found in Standard French and Italian is not the whole story. In Canadian and Maghreb French, Standard Peninsular Catalan, Spanish, Urban Neapolitan and varieties of Rhaeto-Romance spoken in the Engandine Valley the compound past is used only with the auxiliary verb “have”:

  • He llegado [Spanish]
    “I have arrived”
  • Aggiu arrevato [Urban Neapolitan]
    “I have arrived/I arrived”

We also find the inverse type of system, which is unlikely to be familiar to those who've only learnt standard Romance languages. In the Offida dialect, spoken in the Marche region of Central Italy, “be” is used with all classes of main verb:

  • so dÉ™rmitÉ™ [Offida]
    “I've slept/I slept” (literally, I am slept)
    b. so vÉ™'nutÉ™
    “I've come/I came” (literally, I am come)

So far so simple? Perhaps. But there is yet more variation to be found. All the languages we've seen so far make use of a helping auxiliary verb which is decided by the type of main verb: verbs like arrive may take a different auxiliary to verbs like speak. If we look only at standard Romance languages we would think that all variation is determined in this way. In fact, we find a number of lesser-known languages which reveal even more intriguing patterns.

Consider Ariellese, a language spoken in Eastern Abruzzo in Central Italy. Here we have a system which at first appearances looks very similar to French and Italian. But appearances can be deceptive. We find, just like French and Italian, that either the verb for “have” or “be” can appear with the past participle of the main verb. But the type of main verb is not what decides whether you choose “have” or “be”. Instead this choice is determined by the type of subject. When the subject is “I”, “you”, “we” or “you all”, “be” is chosen. When the subject is “he/she/it” or “they”, “have” gets used:

  • So magnate/'rrevate
    “I have eaten/arrived” (literally: I am eaten/arrived)
  • Si magnate/'rrevate
    “You have eaten/arrived” (literally: you are eaten/arrived)
  • A magnite/'rrevate
    “He/she/it has eaten/arrived”
  • Seme magnite/'rrivite
    “We all have eaten/arrived” (literally: we are eaten/arrived)
  • Sete magnite/'rrivite
    “You all have eaten/arrived” (literally: you all are eaten/arrived)
  • A magnite/'rrivite
    “They all have eaten/arrived”

Romanian presents us with yet another type of system where the choice of auxiliary a fi “be” over a avea “have” indicates a degree of doubt on the speaker's part as to whether the event took place:

  • Maria a râs toată seara
    “Maria has laughed all evening”
  • Nu cred să fi plecat
    “I don't think that she has left” (literally: I don't think that she is left)

Taken alongside each other Ariellese and Romanian tell us something important. What may look on the surface to be a pattern we are already familiar with can have quite different rules underlying it from one language to the next.

Considered together, the patterns we've observed here show very clearly that within their grammar Romance languages can differ more from each other than we might think. This small taster of the variation we find also demonstrates that to find a treasure-trove of different language structures we might think of as unusual or exotic, we need only look to the languages we find on our doorstep.

Back to blog