Music: Practice & Theory Stack Exchange is a question and answer site for musicians, students, and enthusiasts. It only takes a minute to sign up.
Sign up to join this community
In learning the guitar I have run across various time signatures. In each time signature it seems as if you are supposed to emphasize different beats. Is this something that you just have to intuit, is there a formula for this, or is it just convention?
For example:
There is no clear and fast rule. The emphasis is entirely dependent on the rhythm of the tune. While there are some standard feels for different time signatures, they are broken as often as they are kept. In terms of standard feels, I'd say that all your examples are correct. It's important to note that 5/4 with accents on the 1 and 4 feels just like 3 + 2 (ala Take 5) and 6/8 generally feels like 2/4 with triplets instead of 1/4 notes (like House of the Rising Sun).
But if you listen to anything that really f&*ks with time signatures, these rules will fall apart quickly. Listen to Schism by Tool, for instance. The main riff goes back and forth between 5/8 and 7/8, but it's just a triplet and then 1/8th notes. The accent seems to roll between the beats. And then it changes to 6/8 and 7/8 and the feel of the piece keeps. Eventually the tune ends in 4/4 with one of the most rolling, odd sounding 4/4s I've heard in a while. After 5 minutes in odd signatures, it's almost hard to believe that the end is in the "common" time signature.
So basically, yes, you've got the standard down. But all the interesting stuff happens on the edges by bucking the rules. Even the standard 5/4 piece, Take Five, was revolutionary because no one had really done that before. It's only "standard" now because Brubeck paved the way.
What are you are referring to is called harmonic stress. It is the metrical stress accorded to a pulse based on its position within a grouping.
It should be noted that, while sometimes musical performers add accentuation to notes or chords occurring on strong harmonic stresses, this is not always the rule. The strength of the beat is actually not a function of amplitude or musical performance, but of the perceived importance of that pulse within a larger grouping.
As you note, the downbeat (the first beat in a measure) is almost always the strongest harmonic stress. Actually, it is not as simple as having just stressed or not-stressed beats. There are degrees.
In common time, the stress is STRONG WEAK strong weak. (1 and 3 are strongest, 2 is weak and 4 is weakest.)
In waltz time, the stress is STRONG WEAK weak.
In cut time the stress is STRONG WEAK.
In odd time meters, the stress is usually just a direct derivative of one of the above. However, I think your idea about the 5:4 stress is not quite complete. It may be possible to have a measure of 5 that is in the pattern of 4+1 or 1+4 as you mention, but I believe 5:4 is usually either 3+2 or 2+3. This affects both how it is notated and how the harmonic stress is heard. So it's either STRONG WEAK weak STRONG WEAK (as would occur from combining a measure of three with a measure of two) or STRONG WEAK STRONG WEAK weak (as would occur from combining a measure of three and a measure of two.)
7/4 time would be seen as a combination of 4+3 or 3+4, and would inherit the stresses that follow.
11/4 would have to be either 4+4+3, 4+3+4 or 3+4+4, but this is a situation we rarely find ourselves in, isn't it?
As for compound triplet meters such as 6/8, 9/8 and 12/8, they are really just measures of two primary beats, three primary beats, or four primary beats, and would have pretty much the same harmonic stress as a measure of 2, 3 or 4 would in the above.
As for notes not occurring within the pulse, such as the "and of 1", they would have the weakest of any harmonic stress, and would also tend to be perceived as inheriting any of their harmonic stress from the pulse which they are subdivided from. So the "and of 1", if it had any real harmonic stress, would be thought of as having the harmonic stress of 1. Indeed, if the measure began with an eighth note rest, the "and of 1" may possibly be heard as an event on beat 1 that was late to enter. However, worrying about the potential metrical stress of subdivided notes is probably not necessary to have a strong grasp of this idea, so we should probably leave it at the main pulses.
The time signature has no bearing on whether a beat may be accented or not. Accents on the weak part of the beat can happen regardless of the time signature. This happens often and is called syncopation.
