Just like a speaker, sound sources get quieter the further away they are from whatever is listening.  In this case we're referring to microphones  The 6dB/doubling of distance rule we talked about above (also known as the inverse square law) becomes very critical in the space between a handheld microphone capsule and the vocalists mouth.  If the microphone is held 8 inches from their mouth (like a news reporter), the sound at the microphone would be 12 decibels quieter than if it were held 2 inches from their mouth (2 inches doubled to 4 inches = -6dB; 4 inches doubled to 8 inches = another -6dB).  Since it's the sound engineer's role to balance the sound levels with the mixer, their job becomes much more challenging if the distance between the mic and the mouth is not kept consistent throughout the performance.  To put this in context, one press of your TV remote "volume" button is roughly 1dB difference.  So a person talking, especially an animated one, could have a fairly wildly varying range of amplitude without careful attention to how the audio is processed by the engineer, and some basic microphone technique by the presenter or performer.

When someone steps back from a microphone and it gets too quiet for all of the listeners to hear them, the sound engineer (who is paying close attention to these details) will boost the level of that mic signal.  When the same person then steps closer to the microphone, the level is immediately increased and the sound engineer has to reduce that same level for the sound in the house system to stay the same.  For a consistent, professional sounding mix, it's important to be aware of how you are holding the mic and keep it as consistent as possible for the duration of the performance. Also remember that sound can transmit through surfaces, so tapping your finger on the microphone handle will be amplified through the system and should be avoided so that it doesn't distract the audience from your message.

What about the other sounds entering the mic?  Imagine a lead singer in a band who leaves their mic on stage in the mic stand during an instrumental break.  Typically on small stages, the drummer in the band is positioned directly behind the lead vocalist, and now that microphone is picking up the entire drum set and amplifying it.  A careful engineer will be monitoring these changes to the stage and will adjust each microphone level as needed to best control the sounds entering the mixer.  In microphones, the loudest source at the mic wins, and this goes for foldback monitors as well.  On a typical stage, the vocalist is the quietest instrument acoustically, yet should be the loudest as heard in the mix.  This means the vocal mic usually has more electrical boost than any other mic on the stage, making it very sensitive.  The louder the stage sound (percussion, guitar and bass amps, horns, etc) the more of that background noise gets picked up by the un-muted vocal mic.  For a sound engineer to truly have control over the mix, they want each microphone to pick up as much of the intended source as possible with as little of the background noise as possible.  Imagine boosting a vocal mic in the house system and hearing more cymbals and guitar in the mix as you bring up the vocal channel, and you've just experienced one of the reasons recording studios use isolation booths to separate instruments in a quiet environment.

Now imagine a soft-voiced singer in a band with a loud guitar player.  The singer will probably want their voice quite loud in their monitor so they can hear themselves above the sound of the guitar amp on stage.  The louder the monitor, the more sensitive that microphone becomes, and suddenly the guitar amp is being picked up by the vocal mic and then amplified from the monitor, making it that much tougher to hear the vocals.  In fact, a "hot" vocal mic can actually start to amplify the sound of the monitor itself, which is what leads to that annoying feedback that everyone hates (when the microphone and speaker begin to amplify each other in a cycle, getting louder and louder until something mechanical fails or the microphone level is reduced manually).  

This is where the term "eating the mic" comes from.  If the vocalist keeps the mic right near their mouth, their voice will be significantly louder (as heard by the mic) than the guitar or the monitor, and the monitor level can be lower to get the same amount of vocal.  Keep in mind that the louder the stage volume, the louder the monitors will need to be, which then creates more "stage wash" noise to be overcome by the house system, taking control away from the sound engineer to create a polished mix.  This affect can be worsened in clam-shell performance spaces, corner stages, or anywhere else where hard surfaces are directly inside the overage area of the monitor system.  This is one of the reasons in-ear monitors have become so popular with professional performers.  The in-ear monitors (like an earbud) are too quiet to be heard by the front few rows of the audience or by the microphones themselves, leading to a clearer sound in the house system.  Other techniques to reduce stage volume include drum shields, guitar amp shields, pointing amps away from the audience and vocal mics, and coaching performers to tailor their playing to the size of the venue.