diff --git a/narwhals/_arrow/namespace.py b/narwhals/_arrow/namespace.py
index c7927e2014..a58c595aa8 100644
--- a/narwhals/_arrow/namespace.py
+++ b/narwhals/_arrow/namespace.py
@@ -26,6 +26,7 @@
 from narwhals._expression_parsing import combine_evaluate_output_names
 from narwhals.typing import CompliantNamespace
 from narwhals.utils import Implementation
+from narwhals.utils import get_column_names
 from narwhals.utils import import_dtypes_module
 
 if TYPE_CHECKING:
@@ -161,7 +162,7 @@ def all(self: Self) -> ArrowExpr:
             ],
             depth=0,
             function_name="all",
-            evaluate_output_names=lambda df: df.columns,
+            evaluate_output_names=get_column_names,
             alias_output_names=None,
             backend_version=self._backend_version,
             version=self._version,
diff --git a/narwhals/_dask/namespace.py b/narwhals/_dask/namespace.py
index 93b02aab4c..5f9fa982f6 100644
--- a/narwhals/_dask/namespace.py
+++ b/narwhals/_dask/namespace.py
@@ -23,6 +23,7 @@
 from narwhals._expression_parsing import combine_alias_output_names
 from narwhals._expression_parsing import combine_evaluate_output_names
 from narwhals.typing import CompliantNamespace
+from narwhals.utils import get_column_names
 
 if TYPE_CHECKING:
     from typing_extensions import Self
@@ -55,7 +56,7 @@ def func(df: DaskLazyFrame) -> list[dx.Series]:
             func,
             depth=0,
             function_name="all",
-            evaluate_output_names=lambda df: df.columns,
+            evaluate_output_names=get_column_names,
             alias_output_names=None,
             backend_version=self._backend_version,
             version=self._version,
diff --git a/narwhals/_duckdb/namespace.py b/narwhals/_duckdb/namespace.py
index 439e22c69b..4ea72fae3b 100644
--- a/narwhals/_duckdb/namespace.py
+++ b/narwhals/_duckdb/namespace.py
@@ -24,6 +24,7 @@
 from narwhals._expression_parsing import combine_alias_output_names
 from narwhals._expression_parsing import combine_evaluate_output_names
 from narwhals.typing import CompliantNamespace
+from narwhals.utils import get_column_names
 
 if TYPE_CHECKING:
     import duckdb
@@ -52,7 +53,7 @@ def _all(df: DuckDBLazyFrame) -> list[duckdb.Expression]:
         return DuckDBExpr(
             call=_all,
             function_name="all",
-            evaluate_output_names=lambda df: df.columns,
+            evaluate_output_names=get_column_names,
             alias_output_names=None,
             backend_version=self._backend_version,
             version=self._version,
diff --git a/narwhals/_pandas_like/namespace.py b/narwhals/_pandas_like/namespace.py
index f369bb3094..4e1b887fc0 100644
--- a/narwhals/_pandas_like/namespace.py
+++ b/narwhals/_pandas_like/namespace.py
@@ -22,6 +22,7 @@
 from narwhals._pandas_like.utils import horizontal_concat
 from narwhals._pandas_like.utils import vertical_concat
 from narwhals.typing import CompliantNamespace
+from narwhals.utils import get_column_names
 from narwhals.utils import import_dtypes_module
 
 if TYPE_CHECKING:
@@ -135,7 +136,7 @@ def all(self: Self) -> PandasLikeExpr:
             ],
             depth=0,
             function_name="all",
-            evaluate_output_names=lambda df: df.columns,
+            evaluate_output_names=get_column_names,
             alias_output_names=None,
             implementation=self._implementation,
             backend_version=self._backend_version,
diff --git a/narwhals/_spark_like/namespace.py b/narwhals/_spark_like/namespace.py
index 1b68d0734a..74d0841431 100644
--- a/narwhals/_spark_like/namespace.py
+++ b/narwhals/_spark_like/namespace.py
@@ -17,6 +17,7 @@
 from narwhals._spark_like.utils import maybe_evaluate_expr
 from narwhals._spark_like.utils import narwhals_to_native_dtype
 from narwhals.typing import CompliantNamespace
+from narwhals.utils import get_column_names
 
 if TYPE_CHECKING:
     from pyspark.sql import Column
@@ -51,7 +52,7 @@ def _all(df: SparkLikeLazyFrame) -> list[Column]:
         return SparkLikeExpr(
             call=_all,
             function_name="all",
-            evaluate_output_names=lambda df: df.columns,
+            evaluate_output_names=get_column_names,
             alias_output_names=None,
             backend_version=self._backend_version,
             version=self._version,
diff --git a/narwhals/typing.py b/narwhals/typing.py
index fe04c4703d..9c868694a7 100644
--- a/narwhals/typing.py
+++ b/narwhals/typing.py
@@ -62,6 +62,9 @@ def aggregate(self, *exprs: Any) -> Self:
         ...  # `select` where all args are aggregations or literals
         # (so, no broadcasting is necessary).
 
+    @property
+    def columns(self) -> Sequence[str]: ...
+
 
 class CompliantLazyFrame(Protocol):
     def __narwhals_lazyframe__(self) -> Self: ...
@@ -73,6 +76,9 @@ def aggregate(self, *exprs: Any) -> Self:
         ...  # `select` where all args are aggregations or literals
         # (so, no broadcasting is necessary).
 
+    @property
+    def columns(self) -> Sequence[str]: ...
+
 
 CompliantFrameT_contra = TypeVar(
     "CompliantFrameT_contra",
diff --git a/narwhals/utils.py b/narwhals/utils.py
index da8eadb80e..9d286398c7 100644
--- a/narwhals/utils.py
+++ b/narwhals/utils.py
@@ -103,6 +103,10 @@ class _FullContext(_StoresImplementation, _LimitedContext, Protocol):  # noqa: P
         - `_version`
         """
 
+    class _StoresColumns(Protocol):
+        @property
+        def columns(self) -> Sequence[str]: ...
+
 
 class Version(Enum):
     V1 = auto()
@@ -1344,6 +1348,10 @@ def dtype_matches_time_unit_and_time_zone(
     )
 
 
+def get_column_names(frame: _StoresColumns, /) -> Sequence[str]:
+    return frame.columns
+
+
 def _hasattr_static(obj: Any, attr: str) -> bool:
     sentinel = object()
     return getattr_static(obj, attr, sentinel) is not sentinel