diff --git a/pyadjoint/optimization/optimization.py b/pyadjoint/optimization/optimization.py
index 259c6e8b..ec26edcd 100644
--- a/pyadjoint/optimization/optimization.py
+++ b/pyadjoint/optimization/optimization.py
@@ -1,3 +1,5 @@
+from functools import partial
+
 import numpy as np
 
 from ..reduced_functional import ReducedFunctional
@@ -59,8 +61,8 @@ def minimize_scipy_generic(rf_np, method, bounds=None, **kwargs):
     m = [p.tape_value() for p in rf_np.controls]
     m_global = rf_np.obj_to_array(m)
     J = rf_np.__call__
-    dJ = lambda m: rf_np.derivative()
-    H = lambda x, p: rf_np.hessian(p)
+    dJ = lambda m: rf_np.derivative(apply_riesz=False)
+    H = lambda x, p: rf_np.hessian(p, apply_riesz=False)
 
     if "options" not in kwargs:
         kwargs["options"] = {}
@@ -153,8 +155,8 @@ def minimize_custom(rf_np, bounds=None, **kwargs):
     m_global = rf_np.obj_to_array(m)
     J = rf_np.__call__
 
-    dJ = lambda m: rf_np.derivative(m)
-    H = rf_np.hessian
+    dJ = partial(rf_np.derivative, apply_riesz=False)
+    H = partial(rf_np.hessian, apply_riesz=False)
 
     if bounds is not None:
         bounds = serialise_bounds(rf_np, bounds)
diff --git a/pyadjoint/reduced_functional_numpy.py b/pyadjoint/reduced_functional_numpy.py
index bf0d73bf..c3cdc6a2 100644
--- a/pyadjoint/reduced_functional_numpy.py
+++ b/pyadjoint/reduced_functional_numpy.py
@@ -60,12 +60,6 @@ def get_global(self, m):
 
     @no_annotations
     def derivative(self, adj_input=1.0, apply_riesz=True):
-
-        if not apply_riesz:
-            raise ValueError(
-                "ReducedFunctionalNumpy only returns primal gradients."
-            )
-
         dJdm = self.rf.derivative(adj_input=adj_input,
                                   apply_riesz=apply_riesz)
         dJdm = Enlist(dJdm)
@@ -80,12 +74,6 @@ def derivative(self, adj_input=1.0, apply_riesz=True):
 
     @no_annotations
     def hessian(self, m_dot_array, apply_riesz=True):
-
-        if not apply_riesz:
-            raise ValueError(
-                "ReducedFunctionalNumpy only returns primal gradients."
-            )
-
         # Calling derivative is needed, see i.e. examples/stokes-shape-opt
         self.derivative()
         m_copies = [control.copy_data() for control in self.controls]