diff --git i/NEWS w/NEWS
index b30e91d..6ff6e14 100644
--- i/NEWS
+++ w/NEWS
@@ -39,6 +39,7 @@
 
   datamash(1), decorate(1): Fix some minor memory leaks.
 
+  Schedule -f/--full combined with non-linewise operations for deprecation.
 
 * Noteworthy changes in release 1.7 (2020-04-23) [testing]
 
diff --git i/doc/datamash.texi w/doc/datamash.texi
index ac02749..3fcbe79 100644
--- i/doc/datamash.texi
+++ w/doc/datamash.texi
@@ -212,6 +212,9 @@ Skip comment lines (starting with '#' or ';' and optional whitespace).
 @opindex -f
 Print entire input line before op results (default: print only the grouped
 keys).
+Using this option with non-linewise operations has been permitted, but never
+produced very sensible output. Such combinations have been deprecated, and in
+future will result in an error.
 
 @item --group=@var{X[,Y,X]}
 @itemx -g @var{X[,Y,X]}
diff --git i/src/datamash.c w/src/datamash.c
index 25f0fa3..51db417 100644
--- i/src/datamash.c
+++ w/src/datamash.c
@@ -232,6 +232,9 @@ which require a pair of fields (e.g. 'pcov 2:6').\n"), stdout);
       fputs (_("\
   -f, --full                print entire input line before op results\n\
                               (default: print only the grouped keys)\n\
+                            This option is only sensible for linewise operations.\n\
+                            Other uses are deprecated and will be removed in a\n\
+                            future version of GNU Datamash.\n\
 "), stdout);
       fputs (_("\
   -g, --group=X[,Y,Z]       group via fields X,[Y,Z];\n\
@@ -640,6 +643,10 @@ process_file ()
   if (input_header && line_number==0)
     process_input_header (input_stream);
 
+  if (print_full_line && !line_mode)
+    fprintf(stderr, "WARNING: Allowing -f/--full with non-linewise operations \
+is deprecated and will be removed in a future release.\n");
+
   /* If there is an input header line, and the user requested an output
      header line, and the input line was read successfully, print headers */
   if (input_header && output_header && line_number==1)
diff --git i/tests/datamash-sort-header.sh w/tests/datamash-sort-header.sh
index d7bd521..0633184 100755
--- i/tests/datamash-sort-header.sh
+++ w/tests/datamash-sort-header.sh
@@ -57,8 +57,10 @@ B 4" > exp_no_sort_headers ||
   framework_failure_ "failed to write exp_no_sort_headers file"
 
 
-echo "A % 1 1,3
-B ( 2 2,4" > exp_sort_in_header_full ||
+echo "A % 1 1
+B ( 2 2
+A & 3 3
+B = 4 4" > exp_sort_in_header_full ||
   framework_failure_ "failed to write exp_sort_in_header_full file"
 
 echo "GroupBy(x) unique(z)
@@ -71,14 +73,18 @@ A 1,3
 B 2,4" > exp_sort_out_header ||
   framework_failure_ "failed to write exp_sort_out_header"
 
-echo "field-1 field-2 field-3 unique(field-3)
-A % 1 1,3
-B ( 2 2,4" > exp_sort_out_header_full ||
+echo "field-1 field-2 field-3 cut(field-3)
+A % 1 1
+B ( 2 2
+A & 3 3
+B = 4 4" > exp_sort_out_header_full ||
   framework_failure_ "failed to write exp_sort_out_header_full"
 
-echo "x y z unique(z)
-A % 1 1,3
-B ( 2 2,4" > exp_sort_headers_full ||
+echo "x y z cut(z)
+A % 1 1
+B ( 2 2
+A & 3 3
+B = 4 4" > exp_sort_headers_full ||
   framework_failure_ "failed to write exp_sort_headers_full"
 
 
@@ -119,19 +125,19 @@ compare_ out7 exp_sort_out_header ||
   { warn_ "sort-header-out failed" ; fail=1 ; }
 
 echo "$INFILE" | sed 1d |
-  datamash -t ' ' --sort --full --header-out -g 1 unique 3 > out8 ||
+  datamash -t ' ' --sort --full --header-out cut 3 > out8 ||
     framework_failure_ "datamash failed"
 compare_ out8 exp_sort_out_header_full ||
   { warn_ "sort-header-out-full failed" ; fail=1 ; }
 
 echo "$INFILE" |
-  datamash -t ' ' -g 1 --sort --full --header-in unique 3 > out9 ||
+  datamash -t ' ' --sort --full --header-in cut 3 > out9 ||
     framework_failure_ "datamash failed"
 compare_ out9 exp_sort_in_header_full ||
   { warn_ "sort-in-header-full failed" ; fail=1 ; }
 
 echo "$INFILE" |
-  datamash -t ' ' -g 1 --sort --full --headers unique 3 > out10 ||
+  datamash -t ' ' --sort --full --headers cut 3 > out10 ||
     framework_failure_ "datamash failed"
 compare_ out10 exp_sort_headers_full ||
   { warn_ "sort-headers-full failed" ; fail=1 ; }
@@ -157,7 +163,7 @@ printf "" | datamash -t ' ' --sort --headers unique 3 > emp4 ||
 compare_ /dev/null "emp4" ||
   { warn_ "sort+headers on empty file failed" ; fail=1; }
 
-printf "" | datamash -t ' ' --sort --full unique 3 > emp5 ||
+printf "" | datamash -t ' ' --sort --full cut 3 > emp5 ||
   framework_failure_ "datamash failed"
 compare_ /dev/null "emp5" ||
   { warn_ "sort+full on empty file failed" ; fail=1; }
diff --git i/tests/datamash-tests-2.pl w/tests/datamash-tests-2.pl
index d743244..a45ec33 100755
--- i/tests/datamash-tests-2.pl
+++ w/tests/datamash-tests-2.pl
@@ -375,44 +375,21 @@ EOF
 
 my @Tests =
 (
-  # Test 'min' + --full
-  # first, verify test without "--full"
+  # Test 'min'
   ['slct1', '-t" " -g1 min 2', {IN_PIPE=>$in_full1}, {OUT=>"A 3\nB 0\n"}],
-  # Test with "--full", "i2" and "i6" should be displayed
-  ['slct2', '-t" " -f -g1 min 2', {IN_PIPE=>$in_full1},
-    {OUT=>"A 3 i2 3\nB 0 i6 0\n"}],
- # --full with --sort => should not change results
-  ['slct3', '-s -t" " -f -g1 min 2', {IN_PIPE=>$in_full1},
-    {OUT=>"A 3 i2 3\nB 0 i6 0\n"}],
-
-  # Test 'max' + --full
-  # first, verify test without "--full"
-  ['slct4', '-t" " -g1 max 2', {IN_PIPE=>$in_full1}, {OUT=>"A 5\nB 8\n"}],
-  # Test with "--full", "i3" and "i7" should be displayed
-  ['slct5', '-t" " -f -g1 max 2', {IN_PIPE=>$in_full1},
-    {OUT=>"A 5 i3 5\nB 8 i5 8\n"}],
-  # --full with --sort => should not change results
-  ['slct6', '-s -t" " -f -g1 max 2', {IN_PIPE=>$in_full1},
-    {OUT=>"A 5 i3 5\nB 8 i5 8\n"}],
-
-  # Test 'first' + --full
-  # first, verify test without "--full"
-  ['slct7', '-t" " -g1 first 2', {IN_PIPE=>$in_full1}, {OUT=>"A 4\nB 1\n"}],
-  # Test with "--full", "i1" and "i4" should be displayed
-  ['slct8', '-t" " -f -g1 first 2', {IN_PIPE=>$in_full1},
-    {OUT=>"A 4 i1 4\nB 1 i4 1\n"}],
-  # more --full with --sort => see test 'sortslct1' below
-
-  # Test 'last' + --full
-  # first, verify test without "--full"
-  ['slct9', '-t" " -g1 last 2', {IN_PIPE=>$in_full1}, {OUT=>"A 5\nB 3\n"}],
-  # Test with "--full", "i1" and "i4" should be displayed
-  ['slct10', '-t" " -f -g1 last 2', {IN_PIPE=>$in_full1},
-    {OUT=>"A 5 i3 5\nB 3 i7 3\n"}],
-  # more --full with --sort => see test 'sortslct2' below
 
+  # Test 'max'
+  ['slct2', '-t" " -g1 max 2', {IN_PIPE=>$in_full1}, {OUT=>"A 5\nB 8\n"}],
 
+  # Test 'first'
+  ['slct3', '-t" " -g1 first 2', {IN_PIPE=>$in_full1}, {OUT=>"A 4\nB 1\n"}],
+
+  # Test 'last'
+  ['slct4', '-t" " -g1 last 2', {IN_PIPE=>$in_full1}, {OUT=>"A 5\nB 3\n"}],
+
+  #
   # Test --narm - ignoring NaN/NA values
+  #
 
   ## Test with 'NA'
   ['narm1', '--narm sum 1',  {IN_PIPE=>$na1}, {OUT=>"6\n"}],
@@ -664,26 +641,6 @@ my @Tests =
 
 );
 
-
-if ($have_stable_sort) {
-  push @Tests, (
-  # Test 'first' + --full + --sort
-  # NOTE: This is subtle:
-  #       Sorting should be stable: only ordering the column which is used
-  #       for grouping (column 1 in this test). This means that the second
-  #       column (containing numbers) should NOT affect sorting, and the order
-  #       of the lines should not change. The results of this test
-  #       should be the same as 'slct8'. If the system doesn't have stable
-  #       'sort', then the order will change.
-  ['sortslct1', '-s -t" " -f -g1 first 2', {IN_PIPE=>$in_full1},
-    {OUT=>"A 4 i1 4\nB 1 i4 1\n"}],
-  # Test 'last' + --full + --sort
-  # See note above regarding 'first' - applies to 'last' as well.
-  ['sortslct2', '-s -t" " -f -g1 last 2', {IN_PIPE=>$in_full1},
-    {OUT=>"A 5 i3 5\nB 3 i7 3\n"}],
-  )
-}
-
 my $save_temps = $ENV{SAVE_TEMPS};
 my $verbose = $ENV{VERBOSE};
 
diff --git i/tests/datamash-tests.pl w/tests/datamash-tests.pl
index a9bd53b..d613c93 100755
--- i/tests/datamash-tests.pl
+++ w/tests/datamash-tests.pl
@@ -451,16 +451,14 @@ my @Tests =
 
   # empty input = empty output, regardless of options
   [ 'emp1', 'count 1', {IN_PIPE=>""}, {OUT=>""}],
-  [ 'emp2', '--full count 2', {IN_PIPE=>""},{OUT=>""}],
+  [ 'emp2', '--full cut 2', {IN_PIPE=>""},{OUT=>""}],
   [ 'emp3', '--header-in count 2', {IN_PIPE=>""},{OUT=>""}],
   [ 'emp4', '--header-out count 2', {IN_PIPE=>""},{OUT=>""}],
-  [ 'emp5', '--full --header-in count 2', {IN_PIPE=>""},{OUT=>""}],
-  [ 'emp6', '--full --header-out count 2', {IN_PIPE=>""},{OUT=>""}],
-  [ 'emp7', '--full --header-in --header-out count 2',
+  [ 'emp5', '--full --header-in cut 2', {IN_PIPE=>""},{OUT=>""}],
+  [ 'emp6', '--full --header-out cut 2', {IN_PIPE=>""},{OUT=>""}],
+  [ 'emp7', '--full --header-in --header-out cut 2',
      {IN_PIPE=>""},{OUT=>""}],
-  [ 'emp8', '-g3,4 --full --header-in --header-out count 2',
-     {IN_PIPE=>""},{OUT=>""}],
-  [ 'emp9', '-g3 count 2', {IN_PIPE=>""},{OUT=>""}],
+  [ 'emp8', '-g3 count 2', {IN_PIPE=>""},{OUT=>""}],
 
   ## Field extraction
   ['f1', '-W sum 1', {IN_PIPE=>$in2}, {OUT=>"5\n"}],
@@ -502,6 +500,22 @@ my @Tests =
   ['mm3', 'absmin 1', {IN_PIPE=>$in_minmax}, {OUT=>"0.0001\n"}],
   ['mm4', 'absmax 1', {IN_PIPE=>$in_minmax}, {OUT=>"-700\n"}],
 
+
+  #
+  # -f/--full
+  #
+
+  # Linewise operations (i.e. MODE_PER_LINE)
+  ['fl1', '-t" " --full cut 2', {IN_PIPE=>$in_g3},
+     {OUT=>"A 3 W 3\nA 5 W 5\nA 7 W 7\nA 11 X 11\nA 13 X 13\n" .
+           "B 17 Y 17\nB 19 Z 19\nC 23 Z 23\n"}],
+  # Non-linewise operations (eg MODE_GROUPBY)
+  ['fl2', '-t" " --full -g3 sum 2', {IN_PIPE=>$in_g3},
+     {OUT=>"A 3 W 15\nA 11 X 24\nB 17 Y 17\nB 19 Z 42\n"},
+     {ERR=>"WARNING: Allowing -f/--full with non-linewise operations " .
+           "is deprecated and will be removed in a future release.\n"}],
+
+
   #
   # Test Grouping
   #
@@ -525,14 +539,6 @@ my @Tests =
   ['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"}],
 
-
-  # --full option - without grouping, returns the first line
-  ['fl1', '-t" " --full sum 2', {IN_PIPE=>$in_g3},
-     {OUT=>"A 3 W 98\n"}],
-  # --full with grouping - print entire line of each group
-  ['fl2', '-t" " --full -g3 sum 2', {IN_PIPE=>$in_g3},
-     {OUT=>"A 3 W 15\nA 11 X 24\nB 17 Y 17\nB 19 Z 42\n"}],
-
   # count on non-numeric fields
   ['cnt1', '-t" " -g 1 count 1', {IN_PIPE=>$in_g2},
      {OUT=>"A 4\nB 3\n"}],
@@ -546,18 +552,21 @@ my @Tests =
      {OUT=>"GroupBy(x) count(y)\nA 5\nB 3\nC 4\n"}],
 
   # Input and output header, with full line
-  ['hdr3', '-t" " -g 1 --full --header-in --header-out count 2',
+  ['hdr3', '-t" " --full --header-in --header-out cut 2',
      {IN_PIPE=>$in_hdr1},
-     {OUT=>"x y z count(y)\nA 1 10 5\nB 5 10 3\nC 8 11 4\n"}],
+     {OUT=>"x y z cut(y)\nA 1 10 1\nA 2 10 2\nA 3 10 3\nA 4 10 4\nA 4 10 4\n" .
+           "B 5 10 5\nB 6 20 6\nB 7 30 7\n" .
+           "C 8 11 8\nC 9 22 9\nC 1 33 1\nC 2 44 2\n"}],
 
   # Output Header
   ['hdr4', '-t" " -g 1 --header-out count 2', {IN_PIPE=>$in_g3},
      {OUT=>"GroupBy(field-1) count(field-2)\nA 5\nB 2\nC 1\n"}],
 
   # Output Header with --full
-  ['hdr5', '-t" " -g 1 --full --header-out count 2', {IN_PIPE=>$in_g3},
-     {OUT=>"field-1 field-2 field-3 count(field-2)\n" .
-           "A 3 W 5\nB 17 Y 2\nC 23 Z 1\n"}],
+  ['hdr5', '-t" " --full --header-out cut 2', {IN_PIPE=>$in_g3},
+     {OUT=>"field-1 field-2 field-3 cut(field-2)\n" .
+           "A 3 W 3\nA 5 W 5\nA 7 W 7\nA 11 X 11\nA 13 X 13\n" .
+           "B 17 Y 17\nB 19 Z 19\nC 23 Z 23\n"}],
 
   # Header without grouping
   ['hdr6', '-t" " --header-out count 2', {IN_PIPE=>$in_g3},
@@ -595,24 +604,20 @@ my @Tests =
   # header-in and header-out => header line should be printed
   ['hdr14', '-t: -H sum 1', {IN_PIPE=>$in_hdr_only},
      {OUT=>"sum(X)\n"}],
-  ['hdr15', '-t: --full -H sum 1', {IN_PIPE=>$in_hdr_only},
-     {OUT=>"X:Y:Z:sum(X)\n"}],
-  ['hdr16', '-t: -s -g1 -H sum 2', {IN_PIPE=>$in_hdr_only},
+  ['hdr15', '-t: -s -g1 -H sum 2', {IN_PIPE=>$in_hdr_only},
      {OUT=>"GroupBy(X):sum(Y)\n"}],
-  ['hdr17', '-t: --full -s -g1 -H sum 2', {IN_PIPE=>$in_hdr_only},
-     {OUT=>"X:Y:Z:sum(Y)\n"}],
-  ['hdr18', '-t: --header-in sum 1', {IN_PIPE=>$in_hdr_only},
+  ['hdr17', '-t: --header-in sum 1', {IN_PIPE=>$in_hdr_only},
      {OUT=>""}],
-  ['hdr19', '-t: -H reverse', {IN_PIPE=>$in_hdr_only},
+  ['hdr18', '-t: -H reverse', {IN_PIPE=>$in_hdr_only},
      {OUT=>"Z:Y:X\n"}],
-  ['hdr20', '-t: --header-in reverse', {IN_PIPE=>$in_hdr_only},
+  ['hdr19', '-t: --header-in reverse', {IN_PIPE=>$in_hdr_only},
      {OUT=>""}],
-  ['hdr21', '-t: -H rmdup 1', {IN_PIPE=>$in_hdr_only},
+  ['hdr20', '-t: -H rmdup 1', {IN_PIPE=>$in_hdr_only},
      {OUT=>$in_hdr_only}],
-  ['hdr22', '-t: --header-in rmdup 1', {IN_PIPE=>$in_hdr_only},
+  ['hdr21', '-t: --header-in rmdup 1', {IN_PIPE=>$in_hdr_only},
      {OUT=>""}],
-  ['hdr23', '-t: --header-in rmdup 1', {IN_PIPE=>""}, {OUT=>""}],
-  ['hdr24', '-t: -H rmdup 1', {IN_PIPE=>""}, {OUT=>""}],
+  ['hdr22', '-t: --header-in rmdup 1', {IN_PIPE=>""}, {OUT=>""}],
+  ['hdr23', '-t: -H rmdup 1', {IN_PIPE=>""}, {OUT=>""}],
 
   # percentile operation has special header handling (which includes
   # the percentile value).
@@ -624,8 +629,6 @@ my @Tests =
   # Test single line per group
   ['sl1', '-t" " -g 1 mean 2', {IN_PIPE=>$in_g4},
      {OUT=>"A 5\nK 6\nP 2\n"}],
-  ['sl2', '-t" " --full -g 1 mean 2', {IN_PIPE=>$in_g4},
-     {OUT=>"A 5 5\nK 6 6\nP 2 2\n"}],
 
   # Test countunique operation
   ['cuq1', '-t" " -g 1 countunique 3', {IN_PIPE=>$in_g3},