Sustaining long-term commitment through Karmic action • Adapting's Blog

1. Karma is not necessarily what you think

The concept of Karma has a rather rational interpretation: it’s non other than the law of cause and effect. Good actions bring about good outcomes. However, there’s an important subtlety that sets the stage for the more mystical interpretation we are used to.

Outside of specific domains, in the general chaos of ordinary lives, there are so many factors at play that we can hardly keep track of everything. As a result, some efforts look wasted (“effectless causes” as it were) and some outcomes seem to come out of the blue (”causeless effects”).

2. Building a compass against cloudy skies

In this context, the idea of karmic action is to focus on making the right kind of efforts, disregarding results as they seem to come in so many unexpected ways. Of course, the right kind of efforts are those that generally bring about good outcomes (hence the importance of understanding causal relationships), but in any particular pursuit, it’s useful to surrender expectations and focus on accumulating good Karma.

This helps to concentrate your energy on your actual actionables and protects motivation from the demoralising effect that high enough expectations¹ tend to have.

3. Using the compass

The tricky thing is that the mind likes to be in the future (when it’s not in the past) and it’s always coming with all kinds of stories about how things will turn out. With your next action, what are you trying to achieve? Can you abandon your expectation and act from the conviction that it’s just the right thing to do?

It may sound kind of stupid, but when you do something just for the sake of doing it (of having decided to do it), you can do anything you want for as long as you need (material limitations aside). This is the path towards building intrinsic motivation, which is critical to sustain commitment towards long-term goals.

Bottom line

In summary, good causes bring about good effects, but this does not happen consistently due to the complicated web of factors involved. Thus, it’s best to first study cause-effect relationships, to identify the kind of actions we want to cultivate, and then in any particular endeavour act with no expectation.

Addendum for the quantitatively inclined - Karma to filter out the noise

Consider a typical linear regression (could be non-linear) where:

$y$ is your variable of interest (your desired effect, like a good grade, a high company valuation or a high effectiveness in a medical drug);
$\mathbf{x}$ is a vector containing the variables you can control (your causes);
$A$ a matrix representing the causal linkage of these causes (how much they impact on your target variable);
$\epsilon$ a gaussian (WLOG) random noise due to factors you cannot control.

y = A \mathbf{x} + \epsilon, \quad \epsilon \sim \mathcal{N}(0, \sigma)

Imagine 50% of the variability of $y$ is due to $A\mathbf{x}$ and the other 50% is due to $\epsilon$ . If you focus your attention on $y$ , half of the time you’ll be chasing noise² and you’ll be wasting your time and effort. Focusing on $\mathbf{x}$ instead is twice more productive.

Of course we ultimately care about $y$ , and the fact that half of its variance is driven by noise may be an issue. This is where the step of causal discovery comes in, we may want to stop periodically and attempt to include more causes into our model so that we have more levers to work with. This is basically a restatement of the process discussed in the body, but expressed from a quantitative perspective.

Footnotes

As they say, expectations have the power of making any reality, no matter how extraordinary it is, disappointing. If you think about it, we often lose motivation because we are not getting the results we thought we would. Note this is quite distinct from not having made any progress or the work not being meaningful. ↩
Technically one cannot jump from variance (50% variability) to frequency (half of the time), but you get what I’m saying; it’s just a figure of speech for a high level example. ↩