In the previous video we've learned that related languages usually look similar to each other. Does this mean that languages are always related if they share words that look alike? Let's take a look at the words for 'car' in a number of different languages. As you can see, these words are similar in both meaning and form. Does this mean that all these languages descend from a common ancestor? You no doubt realize that this does not need to be the case. We know that the car is a fairly recent invention so we can assume that with the introduction of the current worldwide, the corresponding word was also borrowed into many different languages. Because borrowings always look alike, we cannot use them to determine genetic relationships. Instead, we should rather look at basic vocabulary that is not so easily borrowed such as words for body parts, kinship terms, words for native animals, and verbs. If we do this for the languages that we've just seen, for example, by comparing words meaning 'nose', it's immediately clear that they're not so alike at all. If you look more closely however, you will see that some of these words for nose do look rather similar to each other. In this case, we are in fact dealing with languages from three different language families including Indo-European and one language that is unrelated to all the others, namely Basque. But even in basic vocabulary, we shouldn't put too much weight on single words. Take this example, Mbabaram is a now extinct language that used to be spoken in Northern Australia. The Mbabaram word for 'dog' was dog, exactly the same as in English. Does this mean that English and Mbabaram are related to each other? Not necessarily. Only one word that seems to be shared is not sufficient to establish a genetic relationship. Especially for such short words, there is a distinct possibility that their similarity is due to chance. Compare the following list of words from Sanskrit, Greek, and Latin with our English translations. We see the words for 'star', 'father', 'foot', 'knee', and 'night'. Many of these words show similarities across these languages which means that we're not dealing with chance resemblances. Moreover, these words are also part of the verb basic vocabulary. It is therefore very unlikely that they're all borrowed. We can conclude that these languages all inherited these words from a common ancestor, or, in other words, that they are genetically related to each other. Another clue for determining genetic relationships can be found in similarities in inflection. Grammatical markers will often be very similar between related languages. In Indo-European languages for example, we see clear resemblances between the endings of nouns. The ending is the bit at the end of the word that contains information about number and about the function of the word in a sentence. For example, whether it's the subject or object of a sentence. The shape that a noun takes to express such a function is called a case. The nominative case ending, for example, indicates that the noun is the subject of a sentence. You may be familiar with the language that differentiates between different cases in the noun if you know some Latin or German. Here you see the noun for "cloud" in Sanskrit, Hittite, Old Church Slavonic and Greek, and it's different cases in the singular. The nominative singular is shorter than the others and in Greek and Old Church Slavonic it shows an exceptional o-vowel in the second syllable. In the genitive and dative locative singular all languages show similar endings. Similarly, verbs in these different languages show very clear similarities. Here you see the paradigm of the present tense of the verb meaning "to go" in five Indo-European languages. Some of the recurring elements are marked red. A few remarkable parallels are the change of the initial vowel between the singular and the plural in Sanskrit and Greek and the different structure of the third person plural form when compared to the other plural forms in Sanskrit, Hittite, Greek, and Latin. This last example, the occurrence of shared irregular alternations in the same grammatical category, is a very important indication of a genetic relationship. Such irregularities can often be found in the pronoun. In English for example, the word "me" is used as the object form of "I". This is an irregularity. Within English there's no way to predict on the basis of "I" that the corresponding objects would be "me". In Latin we also find an irregularity in the first-person pronoun. The direct object form of "ego" is "mē". It is striking that the Latin pair "ego mē" is very similar to English "I and me". What's more, we see the same kind of irregularity in Sanskrit, "ahám, mā " and Greek, "egṓ, (e)mé". It's highly unlikely that these parallel irregularities all arose by chance. Therefore, this pronoun on its own is already a very strong piece of evidence that these languages are related to each other. How do you determine whether two languages are genetically related? First, look for a sizable amount of similar words in the basic vocabulary of language. Try not to include any loan words. Secondly, look for structural similarities in languages, for example, in inflectional patterns. Finally, look for shared irregularities in the grammar. If two languages show the same irregularity that you cannot explain on the basis of the grammatical rules of the languages themselves, they likely both inherited the irregular feature from the same common ancestor. If you find such similarities, you're dealing with related languages. That was it for this video, thank you and see you in the next.
