A power supply's current rating is telling you how much current it can supply, not how much current it does supply. It's actual output current is determined by whatever you connect to it, called the load. That current rating is only a maximum, so if you have a supply rated at "12V, 2A", and you connect a human body across it with a resistance of 100kΩ, the current that will pass through the body can be calculated by Ohm's law:
$$ I = \frac{V}{R} = \frac{12V}{100k\Omega} = 120\mu A $$
That's way less than the 2A that the supply is capable of. If you wanted the entire 2A to flow, you would need a load of:
$$ R = \frac{V}{I} = \frac{12V}{2A} = 6\Omega $$
So, if the current required to kill you is 10mA, you would need your body to have a resistance of:
$$ R = \frac{V}{I} = \frac{12V}{10mA} = 1.2k\Omega $$
In other words, if you place your left hand on one terminal of a 12V battery, and the right hand on the other, and the total electrical resistance between the two teminals (consisting of terminal-to-skin resistance, and the resistance from left hand to right via your hands, arms and torso) is 1200Ω, the resulting current through your arms and torso will be 10mA, and you will die, possibly. This is a very simplistic model, though.
It's very rare (though not impossible) to find such a low body resistance. It would normally be in the tens or hundreds of kilohms, meaning that 12V batteries are not considered to be dangerous. Even 48V would probably not be enough to do any harm, since the determining factors are both voltage and body resistance, not voltage alone, and body resistance is generally too high to draw any harmful amount of current from the supply.
This doesn't mean you won't feel anything. The various nerves in your body (particularly near the skin) are sensitive to potential (as opposed to flowing current), and any voltage difference will stimulate them, resulting in pain and muscular spasm. So even if your body resistance is high enough to avoid lethal currents, you can still get a "shock" and the associated spasms.
The heart's rhythm can easily be upset by potential differences messing with the potentials that are supposed to be there. This can occur even if the duration of voltage applied to the body is very short, too short to cause any physical damage. Voltage is a killer in that respect.
Physical damage (burning) occurs through heating, which is related to power being dissipated in the body. Since the current is the same along its entire path through the body, heating due to \$P = I^2 \times R\$ power dissipation is greatest where resistance is greatest, which is usually at the skin. Heating will continue for as long as current flows, so where a short high-voltage shock can stop the heart, burn severity will depend on the duration of the shock, and the current flowing. In this sense, it's current that kills.
As you can see, any idea that it's voltage that kills, or current that kills, is over-simplistic. I suppose you could generalise that the culprit can be either; voltage upsets the nervous system, and current burns, both of which can be fatal. However, current depends on body resistance and voltage, so you could argue that it's neither current nor voltage, it's actually hydration and electrolytes that kill.
I feel comfortable with voltages up to 60V or so. I will happily touch AC or DC voltage sources up to that point, though I would never knowingly touch any voltage source over 30V with wet hands, since clearly that will increase current flow. I will have no problem touching a 10A 24V supply, because that will not cause 10A to flow through me.