I'm afraid there isn't a way to circumvent the comparisons, or get rid of any of them. All have to be tested to ensure the match.

If it is that you just need to get rid of specifically the "AND" words, or need a shorter form of writing it, then certainly row-values is the way.

Example:

SELECT K1,K2,K3,K4 FROM T1,T2,T3,T4 WHERE (K1=K2) AND (K1=K3) AND (K1=K4)
vs.
SELECT K1,K2,K3,K4 FROM T1,T2,T3,T4 WHERE (K1,K1,K1) = (K2,K3,K4)

Note that it still compares all, so you did not save any CPU cycles (in fact, it may take more CPU cycles to do this, I'm not sure), but you did save some characters worth of human typing.

That should work just fine, what are the parameters that go into ?1 and ?2 ??

Here are examples of the principle working, using your tables:

-- Reference Query:
SELECT FIRST,LAST,NOTESA,NOTESB FROM T1,T2,T3,T4 WHERE (T1K1=T2K1) AND (T1K1=T3K1) AND (T1K1=T4K1);

  -- FIRST| LAST|NOTESA   |NOTESB   
  -- -----|-----|---------|---------
  --  ONE | TWO |NOTEA 1  |NOTEB 1  
  --  ONE | TWO |NOTEA 2  |NOTEB 2  
  --  ONE | TWO |NOTEA 3  |NOTEB 3  
  --  ONE | TWO |NOTEA 4  |NOTEB 4  


-- Select by Row-Values:
SELECT FIRST,LAST,NOTESA,NOTESB FROM T1,T2,T3,T4 WHERE (T1K1, T1K1, T1K1) = (T2K1, T3K1, T4K1);

  -- FIRST| LAST|NOTESA   |NOTESB   
  -- -----|-----|---------|---------
  --  ONE | TWO |NOTEA 1  |NOTEB 1  
  --  ONE | TWO |NOTEA 2  |NOTEB 2  
  --  ONE | TWO |NOTEA 3  |NOTEB 3  
  --  ONE | TWO |NOTEA 4  |NOTEB 4  



-- Select by row-values and parameter specifier:
SELECT FIRST,LAST,NOTESA,NOTESB FROM T1,T2,T3,T4 WHERE (T1K1, T2K1, T3K1) = (T2K1, T3K1, T4K1) AND T1K1 <= 2;

  -- FIRST| LAST|NOTESA   |NOTESB   
  -- -----|-----|---------|---------
  --  ONE | TWO |NOTEA 1  |NOTEB 1  
  --  ONE | TWO |NOTEA 2  |NOTEB 2  

Having said all that, I'm agreeing with Gunter, in joins, specifying the on-clauses for each bind feels like a much more intuitive way and will improve future understanding of the next person to read your code.

SELECT K1,K2,K3,K4 FROM T1,T2,T3,T4 WHERE (K1,K1,K1) = (K2,K3,K4)

I've never seen a WHERE clause expressed like this! And it works!

I tried:

select * from employees where (lastname,state)=('Adams','AB');

This has values on one side of the equality condition; this works also.

Implicitly, each atomic test is ANDed.Is there a construction or ORing?

Is SQLite's dialect of SQL at this level of detail and including esoteric constructions such as the one above documented somewhere?

Of course it is...
See here: sqlite.org/rowvalue.html

It's not very unique or esoteric actually, and also not simply for making more fun WHERE clauses.

Much of its power lies in the ability to ASSIGN values in this way.

Consider a query like:

UPDATE sales SET (customer, item, qty, price) = ('John', 'Waffle', 2, 2.44);

or more elaborately:

UPDATE games SET (publisher, title, release_date) = (
  SELECT A.publisher_name, B.name, B.release
    FROM publishers AS A
    JOIN pub_titles AS B ON B.publisher = A.id
   WHERE A.short_name = 'Konami'
)
 WHERE id = 44;

Prior to row-values, that would have required 3 different queries in SQLite.