This is a (rather free) translation of an article by Abdallah el Mountassir, entitled Comparer, différencier: l’expression de la comparison en berbère (tachelhit) du sud-ouest marocain, in: Faits de langues, n°5, Mars 1995. La comparaison. pp. 99-107.
This is well worth the read, as it covers an essential part of everyday communication. If you can read French, I recommend reading the original. I’ve merely translated this as part of my effort to disseminate as much learning material to as many people as possible.
Sidi el Mountassir, thank you and bless you for making the world rich in works on Tachelhit which are modern, academic, and accessible.
While the Berber language does not have specific terms to state comparison, the absence of such terms does not mean that the notion of comparison is inexistent. The idea of “to compare” finds its equivalent in the Tachelhit verb “to differentiate, to find distinction between”, sna7ya. When you want to compare, for example, two objects, you say sna7ya-tn, “contrast them”. You have then to determine in what ways they are different: weight, size, color, volume, price, etc. When two objects are different, you say na7ya-tn, “different-them”, which can also be translated as “different but comparable”. If the objects are different, that means that they are not identical (or equal): ur-tn saswa (neg-they.similar), “they are not similar”.
So we have two verbs which enter into an opposition: na7ya “be different” and saswa “be similar”. It is from these verbs that Berber constructs comparative relationships. People, objects, properties can be either identical (saswa) or different (na7ya). In other words, we have here a binary system which rests on two essential comparative relations: likeness and difference. The relation of difference corresponds to two other relationships: inferiority and superiority.
Comparative relationships of difference
As we have just seen, Berber does not have specific formal markers to express these two comparative relations of likeness and difference. The language resorts to lexemes. Entities and events are not treated in the same manner:
A) Comparison of entities: If we are contrasting two or more entities, we have recourse to two verbs: ati “surpass (in number, quantity, size)” and af “be better”. These two verbs ati and af introduce respectively a comparison of quantity and of quality. Take a look at these two statements:
1) tuti lxdmt noBrahim ti n-Hmad
she.exceeds work of-Brahim that of-Hmad
Brahim has more work than Hmad
2) tuf lxdmt n-Brahim ti n-Hmad
she.better work of-Brahim that of-Hmad
The work of Brahim is better/more interesting than the work of Hmad
The entity “work” is thus evaluated here from a quantitative point of view with the verb ati, and from a qualitative point of view with the verb af. It is important to note that these two verbs are not used are not used except in a comparative structure. Their use in a context always necessitates the presence of compared and comparer. The semantics of these two comparative verbs always implies the idea of “surpassing, superiority to”. They express a relation of superiority. To account for the meaning of the statements 1 and 2, we can paraphrase them thus:
1) “the quanitity of Brahim’s work is surpassed by the quantity of Hmad’s work.”
2) “the quality of Brahim’s work is surpassed by the quality of Hmad’s work”
While Berber possesses these two verbs of superiority, it lacks any terms which mean “to be inferior”. One can never say in this language “less than” or “less good”. Berber doesn’t have a direct way to express inferiority. Instead, we must use the comparative of negative superiority. So we say:
1) ur tuti lxdmt n-Hmad ti n-Brahim
neg. she.surpasses work of-Hmad that of-Brahim
≃Hmad has less work than Brahim
2) ur tuf lxdmt n-Hmad ti n-Brahim
neg. she.better work of-Hmad that of- Brahim
the work of ≃Hmad is less interesting than the work of Brahim.
B) Comparison of events: We distinguish two types of verbs in Berber: verbs of action and verbs of state. This distinction is important to note here for, as we are about to see, the Berber comparative system uses different methods depending on the type of verb in question.
By verbs of state, we often mean verbs which denote physical or moral qualities: colors, dimensions, weight, age, spiritual activity, etc. These verbs are characterized, in formal structure, by an initial vowel i-: imim, to be soft, izDuy, to be heavy, ighzif, to be big, imlul, to be white, isdid, to be thin, izur, to be fat, ifsus, to be light, etc.
Before examining the characteristics of this category of verbs, we should first give some clarifications concerning the verbal system in Berber. This system is aspectual, rather than temporal, with two fundamental units corresponding to the opposition of completion/incompletion. The incomplete usually expresses the process in course of action, or a habitual process. The complete describes the process as achieved and definite. From the verb shsh “to eat”, we can have the two following statements:
3)ar ishtta Brahim islman
incomp. he.eats Brahim fish
Brahim is in the process of eating fish
4) ishsha Brahim islman
he.eats comp. Brahim fish
Brahim ate fish
While this construction is valid for the type of verbs indicating a process like azzl “run”, su “drink”, ara “write”, amz “grasp”, etc., it is not the case of these verbs of state/quality such as: imim “be soft”, ifsus “be light”, rgh “hot”, etc.
5) ar irqqa l7lib
incomp. hot milk
the milk is currently hot
6) irgha l7lib
comp. hot milk
the milk is hot
The conceptual relation which links statements 3 and 4 is different from that which links 5 and 6.
In the first case, the incomplete in phrase 3 describes the process in course of action and the complete in phrase 4 is presented as achieved and finished. Note that between the two processes, there is a complete divide. In the second case, the incomplete/complete opposition brings into play different conceptual operations.
So, the completeness of the verb in phrase 6 indicates the result of the process expressed in phrase 5. In other terms, the process of heating in 5, ar irqqa, resulting logically in a state of heat irgha (6). This state comes thus necessarily after the accomplishment of a process. There is thus a continuity between the two processes.
Now we distinguish two types of verbs corresponding to two semantic classes: a class of verbs marking a process like azzl “run”, ftu “leave”, etc., and a class of verbs of state-quality such as rgh “be hot”, ifulki “be good”, etc. These observations permit us to identify a property which is specific to verbs of quality/state: their gradual/scalable character. In effect, the semantic core of these verbs always evokes the notion of transformation or becoming. If in phrase 5 the process is in the course of realization, it is also in course of transformation: the milk becomes more and more hot, and in phrase 6 there is the maximum degree attained by the temperature of the milk.
It is important to be precise here — in phrase 5, the degree of temperature is oriented toward the highest level. We can flip that around and look at the case where the temperature is oriented toward the lowest level, i.e. that which results in cold, akrram:
7) ar ikkrm l7lib
incomp. cold milk
the milk becomes cold
8) ikrm l7lib
the milk is cold
We thus have two opposing orientations: a course ascending in example 5 and a course descending in example 7. These two meanings can be schematised thus:
This variation of degree is inherent in these verbs forming opposing couples like rgh “hot” / krm “cold”, ighzif “big” / imziy “small”, ifsus “light” / izDuy “heavy” etc. We see these as ‘adjustable’ verbs. These same adjustable verbs, as well as event verbs, admit quantificators (adverbs of degree):
9) ighzzif bahra Brahim
comp. it.big very Brahim
Brahim is very big
10) ar bahra ittxdam Brahim
incomp. much he.work Brahim
Brahim works a lot/too much
The use of the quantifying adverb bahra in these two sentences denotes two opposing semantic properties. In the last sentence, we quantify the work activity of Brahim. For the Berber speaker, this sentence furnishes semantic information which we can paraphrase as: “the quantity of Brahim’s work is big”. It relates to a piece of quantitative information which is in play. As to statement 9, it’s the evaluative dimension which is expressed: the size of Brahim is judged as ‘large’.
These two sentences correspond perfectly to the difference in quantity/quality. From these remarks, we can formulate the following distinction: an evaluation of quality with adjustable verbs, and an evaluation of quantities with non-adjustable verbs. In Berber, the size tighzi is an evaluative property: we postulate an evaluative judgement without giving exact precision. On the other hand, the work lxdmt is apprehended here as a quantifiable property.
Comparative relation and spatial dimension
After having briefly examined the characteristics of adjustable verbs and the opposition of these with non-adjustable verbs, let’s now examine how the difference between these two types of verbs manifests itself at the systemic level of comparison in Berber. In effect, the preceding observations allows us to identify two different comparative structures: the quantitative comparison and the evaluative comparison.
A) Quantitative comparison: In a relationship of difference, the quantitative comparison is associated with a grammatical element uggar “more, large quantity”.
11) ar ittxdam Brahim uggar n-Hmad
incomp. he.work Brahim more of Hmad
Brahim works more than Hmad
This sentence transmits the following semantic content:
-Brahim and Hmad work
-The quantity of Brahim’s work is greater than the quantity of Hmad’s work
-The quantity of Hmad’s work is inferior to the quantity of Brahim’s work.
The last sentence is equivalent to the English: “Hmad works less than Brahim”. But in Berber we can never have this type of sentence. For the speaker of this language, this sense of inferiority rests always in subtext. In other terms, Berber doesn’t have a term which opposes uggar and which could be the equivalent of “less”. It is precisely for this reason that the expression of inferiority is implicit in this language while the superiority is always explicit.
B) Evaluative comparison: With this type of comparison, the language puts other concepts into play, as in the following example:
12) ighzzif Brahim f-Hmad
comp. he.big Brahim over-Hmad
Brahim is bigger than Hmad
What is important to take from this sentence is that the grammatical element f- signifies “above, on”. It’s a locative proposition which we can find in a non-comparative sentence:
13) iskkus Brahim f-lkursi
comp. he.sit Brahim on-chair
Brahim is sat on the chair
The localisator f- appears in its full form flla-, when it is followed by a pronoun. We thus have:
12) ighzzif flla-s Brahim
comp. he.big over-him Brahim
Brahim is bigger than him.
f- and flla- are derived from the term aflla which in Tachelhit means “the top, the superior part”. Any differential relationship with adjustable verbs is conceived from a spatial point of view, in particular according to the dimension high/low. In sentence 12, the size of Brahim is described according to his orientation “toward the top” when compared to Hmad. We can paraphrase the same sentence as such: “the size of Brahim is above the size of Hmad, and the size of Hmad is situated below that of Brahim.” This can be schematized as:
Note that this spatial notion of “high” does not always imply a situation of superiority. We can have a statement which denotes inferiority with this same spatial dimension as in the below example:
14) irxs llimun f-ttfa7
comp. he.cheap oranges over-apples
oranges are cheaper than apples
This sentence tells us that “the low price of oranges is situated above the price of apples”. The semantic content can be schematized as:
What we see in these visualizations is that when confronting two different situations, inferiority and superiority, the language performs the same semantic operation which is the spatial orientation of “top, the superior part of”. All these properties which are conceived in Berber as adjustable: price, distance, age, weight, volume, etc., are conceptualized by the same schema. Knowing also that this language also has the locative preposition ddu “under, beneath” as opposed to f- “on, above”. We might here ask the question: why does Berber not make use of the inverse orientation (descending) of “low, bottom part” to express the case of inferiority? In studying the languages of the world, we see the answer clearly: there is not always a perfect fit between linguistic properties and a given reality.
Comparative relation of similarity
The linguistic expression of similarity rests essentially on two terms: saswa “be similar” and zund “like, as”. These two comparative terms correspond to two types of similarity: complete similarity vs. partial similarity. Let’s compare the two following sentences:
15) saswa Brahim d-Hmad h-tighzi
similar Brahim with-Hmad in-size
Brahim and Hmad have the same size
16) ighzzif Brahim zund Hmad
he.big Brahim like Hmad
Brahim is as big as Hmad
Each sentence presents a different semantic meaning. In sentence 15 there is a complete likeness: the size of Brahim is identical to that of Hmad. On the other hand, in sentence 16, zund “like”, tells us that Brahim and Hmad are big. They are the same height, but without their stature being compared. This sentence can be paraphrased as follows: “Just as Brahim is big, Hmad is equally big.” This partial likeness is expressed by zund “like” which always expresses a subjective judgement specific to the speaker.
The examples given in Berber permit us to note something important: this language only has two comparative relations saswa and na7ya, or sameness and difference, to express the three comparative relationships of superiority, inferiority, and equality. Additionally, in studying the comparative system of Berber, a phenomenon catches our attention: if we observe the current generation of Tachelhit speakers we notice they have a tendency to replace these Berber structures of comparison with those of dialectal Arabic, which, like Berber, is purely oral. However, dialectal Arabic has itself borrowed this type of construction from written Arabic where the methods of comparison are more rich and complex. This explains, without a doubt, that the dearth of specific terms of comparison can be attributed to the oral nature of these languages, and that such terms and structures are developed with writing.