The first movement of Bach’s ‘Ein Feste Burg ist Unser Gott’ cantata is a polyphonic setting of Martin Luther’s chorale melody of the same name. The melodic material and structure (and sometimes even tonality of the phrases) of the chorale melody are reflected in the movement, which is a chorus for choir and orchestra. This is typical of the opening movement of a Bach cantata. Often, as is the case in this chorus, Bach will subject the chorale melody to fugal treatment, the stringency of which varies from cantata to cantata.

The movement opens with an accompanied, decorated statement of the first phrase of the chorale melody in the tonic key of D major. This statement acts as a fugal subject. It is sung by the tenors, doubled by violas and accompanied by cello/harpsichord. At the end of the phrase, the altos sing an answer to the subject. It is a tonal answer, because the melody is modified slightly to avoid the dominant key of A major. The most glaring evidence of this is the persistence of G-naturals, as opposed to the introduction of G-sharps. Note, also, how the melody gravitates towards the note D instead of E; the answer ends on A–D, as opposed to B–E which would have been a direct transposition of the original subject. Tonal answers are very common in fugues. However, even tonal answers are usually harmonised in the dominant key—a defining feature of fugue is, after all, the tonic–dominant entries of the subject. The fact that this passage doesn’t modulate distinguishes it slightly from a true fugue.

Whilst the altos enter with the first phrase of the chorale melody, the tenors proceed to sing the second phrase. This becomes the countersubject. Usually, once the second entry of the subject is complete, there is a brief episode/codetta to modulate back to the tonic in preparation for the third entry. Because no such modulation is necessary here, there isn't one; the third entry comes in straight after the second, in the sopranos. This gives this exposition a rather canonic quality, which is further substantiated by the introduction of a second countersubject in the tenors which is accordingly repeated by the altos at the end of the soprano entry, such that the overall impression is that of an extended melody sung in canon. 

All the voices then assume an ascending melodic shape, which increases tension, or at least creates an audible build-up to a statement of the original, undecorated first phrase of the chorale in the oboes. This statement acts as the cantus firmus of the ensuing passage. The cantus firmus is imitated in canon at the octave by the continuo. Meanwhile, the rest of the voices continue to sing variations of the subject and countersubjects.

This treatment of the chorale melody is used a lot by Bach, both in other cantata choruses but also in chorale preludes for organ. For instance, in BWV 668, the first phrase of the chorale melody ‘Wenn wir in höchsten Nöten sein’ is used as a counterfugue subject. Again, a cantus firmus—this time decorated—version of the melody is played after the exposition, on the upper manual of the organ to make it stand out (similarly to the use of oboe in the cantata). This treatment is repeated for each phrase of the chorale melody—until the final phrase, where the answer is not only inverted, but also diminished! And a little later, just before the cantus firmus entry, there’s also a counter-exposition where the first entry, this time in a different voice, is the inverted version and the second, diminished entry is the original version! 

Coming back to the cantata, towards the end of the cantus firmus statements, there begins in the altos another brief exposition based on the first chorale phrase. This time it starts with the ‘answer’ from the original exposition (i.e. it starts on the note A). The altos then go on to sing the countersubject, but it starts slightly later metrically than in the original exposition (presumably to accommodate for the fact that the answer and subject are now the other way around, with the answer starting on D). After their entry, the sopranos proceed to sing the countersubject as normal, and the exposition progresses as before (albeit with the entries in a slightly different order.) Like before, the second chorale phrase then comes in as the canonic cantus firmus entries in the oboes and continuo.

You can see that unlike BWV 668, the second chorale phrase is introduced by fugal treatment of primarily the first chorale phrase. However, this isn’t inappropriate because the second phrase is highly abundant in the first phrase’s fugal treatment: it’s used as the first countersubject, and is the basis of the second countersubject too. There is plenty of emphasis on it leading up to its entry as the cantus firmus.

Everything heard up to this point is then repeated. This reflects the structure of the original chorale melody, wherein the first two phrases are repeated. 

After this repeat, a new fugal exposition begins using a decorated version of the third phrase of the chorale. This time it starts in the basses, accompanied by cello and harpsichord but not doubled. It opens with a leap of a fifth, from D to A. The overall tonality of the melody is a little ambiguous, though: the leap of a fifth is very tonicising to D major, but the melody ends with a G# rising to an A. In the original chorale melody, this results in mid-phrase modulation. Bach embraces this ambiguity in his cantata setting, adding a chromatic rise in the basses. Perhaps it is a modulating fugal subject, beginning in D major and ending in the dominant key of A major.

The second entry opens with a leap of a fifth from A to E. As previously discussed, this leap of a fifth would typically be modified to a fourth, A to D, to retain an overall key of D major. However, Bach has deliberately utilised a real answer to retain the fifth with its tonicising quality and imply modulation to A major—and yet, because it is, indeed, a real answer, instead of modulating back to D major as is typical practice with a modulating subject, the subject again ends with an added sharp: this time, D#. This implies E minor, seemingly confirmed by the ensuing E–B leap in the alto entry—but this is negated by the subsequent A# implying B minor, which is in turn negated by an E# implying F# minor. Bach never truly settles in any of these keys, bringing the music firmly back to D major as the cantus firmus of the third phrase comes in. The G# in the melody forms a secondary dominant.

Meanwhile, the countersubject seems to be an altered, minor version of the first chorale phrase. The second countersubject could be interpreted as a variation of the second phrase, the ‘ein gute Wehr und Waffen’ melody, given its descending shape and characteristic crotchet rest at the start.

After this ends, an exposition based on the fourth phrase begins. Like the previous one, this is hardly an authentic fugal exposition: Bach similarly cycles through dominant minors as opposed to alternating between tonic and dominant as is typical in fugue. There don’t seem to be any strict countersubjects here, but the first entry of the subject, in the altos, is followed by free counterpoint with similarities to the ‘ein feste Burg ist unser Gott’ melody. The music then modulates back to D major in anticipation of the cantus firmus of the fourth chorale phrase.

The exposition of the fifth phrase also doesn’t follow typical fugal tonic–dominant motion—although the first two entries are a fifth apart. The apparent key is D major, but due to the sparse texture, the fact that the melody seemingly begins on the third scale degree, and the fact that the subject answer is real, the tonality is a little unclear. The third and fourth entries are chromatically altered, minor versions of the subject. The third entry comes in a major sixth above the second entry, and the fourth entry a fourth above the third. The interval of a fourth implies some sort of minor dominant relationship again, but the overall tonality is ambiguous, like in the case of the third and fourth chorale phrase expositions. This time, the cantus firmus passage is in A major, which reflects the tonality of the phrase in the original chorale.

The sixth phrase exposition cycles backwards through dominant minors (or forwards through subdominants) for the first three entries. The third entry is in slight stretto with the second. The fourth entry is a fifth above the third, i.e. the same as the second. This establishes a tonality of F# minor. The cantus firmus passage is in F# minor, reflecting the fact that this is the most explicitly minor phrase in the original chorale melody (although Bach actually harmonises it in E minor in his homophonic settings.)

The final phrase of the original chorale has a similar contour to the first: an overall descending shape. Because of this, Bach reuses the melody he used for the first fugal answer at the beginning of the movement. This part of the cantata is therefore the least representative of the original chorale. It appears to open in B minor, which contrasts the tonic tonality of the original chorale phrase. I think this relative-minor entry of the original subject is somewhat reminiscent of the relative-minor subject entries with which Bach often opens his development sections (such is the case, at least, in the Well-Tempered Clavier.) This adds an element of motivic and structural coherence to the movement, which is otherwise rather compartmentalised in spite of the recurring countersubjects (as is typical of chorale preludes.)

The answer enters in E minor, the subdominant (contrasting typical fugal tonic–dominant motion, but the likes of which is still occasionally found, such as in Cp. X from the Art of Fugue where the answer is also in the subdominant). The answer also enters in slight stretto with the subject, which isn’t the case in the original exposition at the start.

The third entry is also a fourth above the second entry, which might imply cycling through subdominants again, but the naturalisation of D-sharps and introduction of G-sharps establishes the key of A major. The fourth entry, in the altos, is actually the answer from the original exposition—the tonal answer which, as we discussed, is explicitly in the key of D major. The cantus firmus canon comes in with the final chorale melody phrase and a coda ensues. Bach puts down a tonic pedal, but introduces C-naturals to delay the final resolution to D major—something he does very often. To name just one example, observe the ending BWV 925, which is also in D major and also uses a tonic pedal and C-naturals towards the end.

In conclusion, it is clear that the original chorale melody thoroughly permeates the opening movement of the cantata. The movement uses its melodies, structure, and often even tonalities. The individual phrases of the melody are each treated fugally.