Irregular verbs
While many verbs in English follow the relatively simple paradigm illustrated at the beginning of this section, there are some verbs that do not. There are two categories of such verbs:

1. strong verbs (the "transparently irregular")
2. true irregular verbs.

The term "transparently irregular" is sometimes used to describe Jacob Grimm's "strong" verbs that appear irregular at first, but actually follow a common paradigm. This group of verbs is a relic of the older Germanic ablaut system for conjugation. This is generally confined to atypical simple past verb forms, e.g.:

I swim ~ I swam ~ I have swum
I sing ~ I sang ~ I have sung
I steal ~ I stole ~ I have stolen

Another category of "transparently irregular" verbs dates back to Middle English. Some verbs, especially those with a stem ending in an alveolar consonant (/t/, /d/, or /s/), formed a geminate consonant or consonant cluster with the -d suffix. In Middle English, vowels before a consonant cluster often became shorter. As the Great Vowel Shift obscured the connection between long vowels and the corresponding short vowels, transparent irregularities such as the following arose:

I meet ~ I met ~ I have met
I lead ~ I led ~ I have led
I read ~ I read ~ I have read
I lose ~ I lost ~ I have lost
I keep ~ I kept ~ I have kept

True irregular verbs have forms that are not predictable from ablaut rules. The most common of these in English is the verb "to be." A sampling of its verbal paradigm is listed below; the majority of other forms are predictable from the knowledge of these four.
Tense 1st sg. 2nd sg. 3rd sg. 1st pl. 2nd pl. 3rd pl.
Simple present I am You are He/she/it is We are You are They are
Simple past I was You were He/she/it was We were You were They were
Present progressive I am being You are being He/she/it is being We are being You are being They are being

Irregular verbs include eat, sit, lend, and keep, among many others. Some paradigms are based on obsolete root words, or roots that have changed meaning. Others are derived from old umlaut patterns that changes in phonemic structure and grammar have distorted (keep ~ kept is one such example). Some are unclear in origin, and may date back to Proto-Indo-European times.

[edit] Notes

1. In English, a long-standing prescriptive rule holds that shall denotes simple futurity in the first person, and will denotes simple futurity in the second and third persons. In American English, this distinction has largely vanished; will is normally used for both cases, and shall is rare. In British English, adherence to the rule has declined during the 20th century (see Shall and will for a more detailed discussion), although use of shall remains for expressing the simple future in the first person.
2. The distinction between tense, aspect, and mood is not clear-cut or universally agreed-upon. For example, many analysts would not accept that English has twelve tenses. The six "continuous" (also called "progressive") forms in the list above are often treated under the heading of "aspect" rather than tense: the simple past and the past continuous are examples of the same tense, under this view. In addition, many modern grammars of English agree that English does not have a future tense (or a future perfect). These include the two largest and most sophisticated recent grammars:

1. Biber, D., S. Johansson, G. Leech, S. Conrad & E. Finegan. 1999. Longman grammar of spoken and written English. Harlow, Longman.
2. Huddleston, R. & G. Pullum. 2002. The Cambridge grammar of the English language. Cambridge, CUP.

The main argument given by Huddleston and Pullum (pp 209-10) that English does not have a future tense is that "will" is a modal verb, both in its grammar and in its meaning. Biber et al. go further and say that English has only two tenses, past and present: they treat the perfect forms with "have" under "aspect". Huddleston & Pullum, on the other hand, regard the forms with "have" as "secondary tenses".