• Dutch
  • Frisian
  • Saterfrisian
  • Afrikaans
Show all
6.2 Suppletion in the verbal paradigm

We speak of irregularity in case the forms of a verbal paradigm are not completely p[redictable on the basis of a single form. However, most irregular forms do share a similarity among each other. Reconsider the case of boake ‘bake’. It has as reduced stems: bak and buuk. Although these forms are not predictable, they are similar. More specifically, the consonant frame remains constant, and it is just the nucleus (the vowel) which is irregular. This holds true of most irregular verbs. Irregularity targets the stem vowel, but not the consonant frame. There is a small number of cases in which the coda is irregular, that is, the consonant group following the nucleus. In such a case, only the onset is constant throughout the paradigm. This is for example the case with a verb like sjo ‘see’. It doesn’t have a coda in its infinitive, but in the past tense the coda is a velar fricative: saach ‘saw’. It is only a minority of the irregular verbs, in which the coda is affected. But then there is an even higher degree of irregularity, in which some forms of the paradigm of a speicifc verb have nothing in common with each other. This is the case with the most irregular of all verbs in most languages, the verb of being. The 3SG form is ‘is’ has nothing in common with the 1SG form bän ‘am’. Here two completely different stems are involved. And the infinitive weze ‘be’ instantiates a third stem. If there is no similarity between two stems belonging to the same verbal paradigm, we speak of suppletion.


All West Germanic languages testify to suppletion in the verbal paradigm of the verb of being. Suppletion is related to frequency. Only very frequent verbs may exhibit suppletion. The verb of going has an example of suppletion in its paradigm in English: the past tense went is unrelated to the other forms of the paradigm. It derives from a different verb alltogether: Old English wendan ‘turn, go’, which appears in Saterland Frisian as weende ‘turn’. The verb of going in Saterland Frisian is gunge ‘go’. It does not testify to suppletion: all its forms have the same onset before the nucleus: g-. However, the language has developed two other cases of suppletion, which are characteristic of East Frisian and which are not found in the other West Germanic languages. These two cases naturally involve two very frequent verbs, namely the verb of seeing (sjo ‘see’) and the verb of giving (jeeuwe ‘give’). We will discuss both these verbs in some more detail below.

The verb sjo ‘see’ has as its perfect participle the form: blouked. The complete paradigm is given as follows (from J. Hoekstra 2008, on whom the information given here is based).

Table 1: The paradigm of sjo see in Saterland Frisian
1SG 2SG 3SG PL Participle
Present: sjo sjuchst sjucht sjo (säin)
Past: saach saachst saach segen blouked

Suppletion in the paradigm of to see might well be a property of all East Frisian language varieties. It is also found in Wangerooge Frisian:

Table 2: The paradigm of sjo see in Wangerooge Frisian
1SG 2SG 3SG PL Participle
Present: sjo sjuchst sjucht sjo sjoeen
Past: blouked bloukest blouked bloukeden blouked

In Wangerooge Frisian, the suppletive form has extended its domain from the past participle to the simple past. The intrusive form derives from the verb bloukje ‘see’, which reconstructs to be-loukje. It is also interesting that the prefix be- is no longer recognisable, since it is perfectly well able to survive with other verbs. The form loukje is cognate with English to look, West Frisian (now obsolete) loaitsje. There are indications that the paradigm of the verb of seeing was suppletive in other East Frisian varieties as well. The intrusion seems first to have taken place in the form of the past participle. The original participle of sjo is: sain ‘seen’. This form still exists in compounds and derived forms, such as fersjo ‘equip’, which has as its past participle fersain. The verb bloukje ‘look’ does not exist outside the paradigm of sjo ‘see’, where the form of its participle lives on in Saterland Frisian. This is similar to English, where the verb wendan disappeared after its past tense was incorporated into the verb to go. Saterland Frisian didn’t preserve the form loukje ‘look’ either. It was replaced with the Low German form kiekje ‘see’. J. Hoekstra relates the possibility of the intrusion process to semantic factors, which certainly play a role. The second factor is frequency. Suppletion correlates with frequency. Only verbs of very high frequency can be subject to suppletion. The third factor is similarity. It seems that suppletion takes place in case the forms of a very frequent verb resemble each other too much. The effect of suppletion is that it creates dissimilarity among the forms of the paradigm. In this respect, it is interesting that suppletion targets verbs with a monosyllabic infinitive. These monosyllabic verbs have forms which are relatively more similar to each other.

A second example of suppletion involves the Old Frisian verb jeva ‘to give’ (on Old Frisian, see Bremmer 2009 and the references given there). This verb has been completely replaced by the verb reke ‘to hand’. This process may have taken place by suppletion as well. It has even affected the compunds and derived forms, so we find forms like uutreke ‘give out’, and so on. The original stem is preserved in the nominalisation Jeeuwer ‘giver, spender’. In addition, it survives in a couple of derived forms. This subject merits further study.

Finally we point out that there are cases in which two verbs seem to have been collapsed. As a result, one verbal paradigm may have two different and unrelated meanings. A case in point is läite, which means both ‘to let’ (West Frisian litte) and ‘to seem’ (West Frisian lykje). Another example is stete, which means both ‘to bump’ (West Frisian stjitte) and ‘to prick’ (West Frisian stekke).

    printreport errorcite