diff --git i/NEWS w/NEWS
index f769b18..8bb5d43 100644
--- i/NEWS
+++ w/NEWS
@@ -1,5 +1,9 @@
 * Noteworthy changes in release ?.? (????-??-??) [?]
 
+** New Features
+
+  datamash(1): -g/groupby takes ranges of columns (e.g. 1-4)
+
 
 * Noteworthy changes in release 1.8 (2022-07-23) [stable]
 
diff --git i/doc/datamash.texi w/doc/datamash.texi
index e5693a5..e149880 100644
--- i/doc/datamash.texi
+++ w/doc/datamash.texi
@@ -216,8 +216,8 @@ While using this option with non-linewise operations was historically permitted,
 it never produced very sensible output. Such usage has been deprecated, and in a
 future release it will result in an error.
 
-@item --group=@var{X[,Y,X]}
-@itemx -g @var{X[,Y,X]}
+@item --group=@var{X[,Y,Z]}
+@itemx -g @var{X[,Y,Z]}
 @opindex --group
 @opindex -g
 @cindex grouping
@@ -227,6 +227,7 @@ must be sorted by the same fields @var{X[,Y,Z]}. Use @option{--sort}
 to automatically sort the input.
 If @option{--group} is not specified, each operation is performed
 in the entire input file.
+Ranges of field numbers like @var{X-Z} are also supported.
 
 @item --header-in
 @opindex --header-in
diff --git i/src/op-parser.c w/src/op-parser.c
index 6b464e1..cfff7c9 100644
--- i/src/op-parser.c
+++ w/src/op-parser.c
@@ -557,7 +557,27 @@ parse_mode_column (enum processing_mode pm)
     case TOK_INTEGER:
       if (scan_val_int>0)
         {
-          ADD_NUMERIC_GROUP (scan_val_int);
+          int begin_range = scan_val_int, end_range;
+          if (scanner_peek_token () != TOK_DASH)
+            {
+              ADD_NUMERIC_GROUP (scan_val_int);
+            }
+          else
+            {
+              scanner_get_token ();
+              if (scanner_get_token () != TOK_INTEGER
+                  || scan_val_int < (uintmax_t) begin_range)
+                {
+                  die (EXIT_FAILURE, 0,
+                       _("invalid field range for operation %s"),
+                       quote (get_processing_mode_name (pm)));
+                }
+              end_range = scan_val_int;
+              for (int i = begin_range; i <= end_range; i++)
+                {
+                  ADD_NUMERIC_GROUP (i);
+                }
+            }
           break;
         }
       /* fallthrough */
diff --git i/tests/datamash-crosstab.pl w/tests/datamash-crosstab.pl
index e368658..25d9d2c 100755
--- i/tests/datamash-crosstab.pl
+++ w/tests/datamash-crosstab.pl
@@ -171,6 +171,7 @@ my @Tests =
   ['c1','crosstab 1,2 first 3', {IN_PIPE=>$in1}, {OUT=>$out1_first}],
   ['c2','ct 1,2 first 3',       {IN_PIPE=>$in1}, {OUT=>$out1_first}],
   ['c3','ct 1,2 count 1',       {IN_PIPE=>$in1}, {OUT=>$out1_count}],
+  ['c3a','ct 1-2 count 1',      {IN_PIPE=>$in1}, {OUT=>$out1_count}],
 
   # Default operation is count
   ['c4','ct 1,2',               {IN_PIPE=>$in1}, {OUT=>$out1_count}],
@@ -214,8 +215,6 @@ my @Tests =
     {ERR=>"$prog: crosstab supports one operation, found 2\n"}],
   ['e7',  'ct 1:2', {IN_PIPE=>""}, {EXIT=>1},
     {ERR=>"$prog: invalid field pair for operation 'crosstab'\n"}],
-  ['e8',  'ct 1-2', {IN_PIPE=>""}, {EXIT=>1},
-    {ERR=>"$prog: invalid field range for operation 'crosstab'\n"}],
 );
 
 my $save_temps = $ENV{SAVE_TEMPS};
diff --git i/tests/datamash-error-msgs.pl w/tests/datamash-error-msgs.pl
index 34259c1..5c899ef 100644
--- i/tests/datamash-error-msgs.pl
+++ w/tests/datamash-error-msgs.pl
@@ -175,8 +175,6 @@ my @Tests =
   # Invalid field specifications for primary operations
   ['e90',  'groupby 1:2', {IN_PIPE=>""}, {EXIT=>1},
     {ERR=>"$prog: invalid field pair for operation 'groupby'\n"}],
-  ['e91',  'groupby 1-2', {IN_PIPE=>""}, {EXIT=>1},
-    {ERR=>"$prog: invalid field range for operation 'groupby'\n"}],
 
   # values for strbin operation
   ['e92','strbin:-  1',    {IN_PIPE=>""}, {EXIT=>1},
diff --git i/tests/datamash-tests.pl w/tests/datamash-tests.pl
index 687a5b1..d78fb69 100755
--- i/tests/datamash-tests.pl
+++ w/tests/datamash-tests.pl
@@ -323,6 +323,54 @@ my $in_hdr_only=<<'EOF';
 X:Y:Z
 EOF
 
+my $in_multiple_groups=<<'EOF';
+a a a foo
+a a a bar
+a a b baz
+a b b alice
+a b c bob
+b b c eve
+b c d frank
+b c d foo
+b c d bar
+b d d baz
+c d a alice
+c d a bob
+c e b eve
+c e b frank
+c e c foo
+d a c bar
+d a d baz
+d a d alice
+d b e bob
+d b e eve
+e b a frank
+e c a foo
+e c b bar
+e c b baz
+e d c alice
+EOF
+
+my $out_multiple_groups=<<'EOF';
+a	a	a	bar,foo
+a	a	b	baz
+a	b	b	alice
+a	b	c	bob
+b	b	c	eve
+b	c	d	bar,foo,frank
+b	d	d	baz
+c	d	a	alice,bob
+c	e	b	eve,frank
+c	e	c	foo
+d	a	c	bar
+d	a	d	alice,baz
+d	b	e	bob,eve
+e	b	a	frank
+e	c	a	foo
+e	c	b	bar,baz
+e	d	c	alice
+EOF
+
 =pod
   Example:
   my $data = "a 1\nb 2\n";
@@ -457,9 +505,13 @@ my @Tests =
           "line 1 has only 3 fields\n"}],
   ['e25',  '-g 1,,2 sum 1' ,  {IN_PIPE=>"a\n"}, {EXIT=>1},
     {ERR=>"$prog: missing field for operation 'groupby'\n"}],
-  ['e26',   '--collapse-delimiter=foo', {IN_PIPE=>"a\n"}, {EXIT=>1},
+  ['e26',  '-g 4-1 sum 1', {IN_PIPE=>"a\n"}, {EXIT=>1},
+    {ERR=>"$prog: invalid field range for operation 'groupby'\n"}],
+  ['e27',  '-g 1- sum 1', {IN_PIPE=>"a\n"}, {EXIT=>1},
+    {ERR=>"$prog: invalid field range for operation 'groupby'\n"}],
+  ['e28',   '--collapse-delimiter=foo', {IN_PIPE=>"a\n"}, {EXIT=>1},
     {ERR=>"$prog: the delimiter must be a single character\n"}],
-  ['e27',   '-c foo', {IN_PIPE=>"a\n"}, {EXIT=>1},
+  ['e29',   '-c foo', {IN_PIPE=>"a\n"}, {EXIT=>1},
     {ERR=>"$prog: the delimiter must be a single character\n"}],
 
   # No newline at the end of the lines
@@ -504,6 +556,10 @@ my @Tests =
   # Numeric operation on an empty field should not work
   ['f23', '-t: -g1 sum 2', {IN_PIPE=>$in_empty1}, {EXIT=>1},
     {ERR=>"$prog: invalid numeric value in line 1 field 2: ''\n"}],
+  ['f24', '-W -g1,2,3 uniq 4',
+    {IN_PIPE=>$in_multiple_groups}, {OUT=>$out_multiple_groups}],
+  ['f25', '-W -g1-3 uniq 4',
+    {IN_PIPE=>$in_multiple_groups}, {OUT=>$out_multiple_groups}],
 
   # whitespace only, different field delimiters
   ['ws1.1', 'check',     {IN_PIPE=>$ws1}, {OUT=>"1 line, 4 fields\n"}],
@@ -568,7 +624,8 @@ my @Tests =
   # Multiple keys (from different columns)
   ['g8.1',     '-t" " -g1,3 sum 2', {IN_PIPE=>$in_g3},
     {OUT=>"A W 15\nA X 24\nB Y 17\nB Z 19\nC Z 23\n"}],
-
+  ['g8.2',     '-t" " -g1-2 count 2', {IN_PIPE=>$in_g4},
+    {OUT=>"A 5 1\nK 6 1\nP 2 1\n"}],
 
   # count on non-numeric fields
   ['cnt1', '-t" " -g 1 count 1', {IN_PIPE=>$in_g2},